Patent Application
20140309793
The subject of this patent relates to sensing, actuation, and automatic control of physical processes including industrial processes, equipment, facilities, buildings, devices, boilers, valve positioners, motion stages, drives, motors, turbines, compressors, engines, robotics, vehicles, and appliances.
In the foreseeable future, the energy needed to support our economic growth will continue to come mainly from coal, our nation's most abundant and lowest cost resource. The performance of coal-fired power plants is highly dependent on coordinated and integrated sensing, control, and actuation technologies and products.
The implementation of sensors and advanced controls in power systems an provide valuable methods to improve operational efficiency, reduce emissions, and lower operating costs. As new power generation technologies and systems mature, the plant that encompasses these systems will become inherently complex. The traditional process control architecture that includes a conventional process layer, sensing layer, control layer, and actuation layer would no longer be sufficient. In order to manage complexity, the process control architecture that supports the plant control systems need to evolve to manage complexity and optimize performance.
On the other hand, with the advent of information technology, sensor networks have been implemented in more and more industrial plants. Most “modern” sensors and actuators are equipped with Fieldbus, a digital network for the industrial environment, that can send and receive useful information throughout the network. However, much of the information from the sensor networks is not very well utilized due to various reasons.
In the U.S. patent application No. 61/727,045, the entirety of which is hereby incorporated by reference, we described a Self-Organizing Process Control Architecture that comprises a Sensing Layer, Control Layer, Actuation Layer, Process Layer, as well as Self-Organizing Sensors (SOS) and Self-Organizing Actuators (SOA). A Self-Organizing Sensor with an artificial neural network (ANN) based dynamic modeling mechanism to measure a CFB Boiler Bed Height is presented. A method to develop a Self-Organizing Sensor that has one or multiple input variables is disclosed.
In the U.S. patent application, filed on Apr. 15, 2013 and entitled Self-Organizing Multi-Stream Flow Delivery Process and Enabling Actuation and Control, the entirety of which is hereby incorporated by reference, we described a self-organizing multi-stream flow delivery process and the enabling actuation and control system. The method and apparatus of building a general-purpose self-organizing multi-stream flow delivery process are presented. As a case example, an actuation and control system to control a multi-stream liquid flow delivery process using Self-Organizing Actuation and Control Units (SOACU) is described.
In the U.S. Pat. No. 6,684,112, the entirety of which is hereby incorporated by reference, we described a Robust Model-Free Adaptive (MFA) controller for effectively controlling simple to complex processes. The Robust MFA controller provides a wide robust range and can keep the process in control during normal and extreme operating conditions when there are significant disturbances or changes in process dynamics.
First introduced in 1997, the Model-Free Adaptive (MFA) control technology overcomes the shortcomings of traditional Proportional-Integral-Derivative (PID) controllers and is able to control various complex processes that may have one or more of the following behaviors: (1) nonlinear, (2) time-varying, (3) large time delay, (4) multi-input-multi-output, (5) frequent dynamic changes, (6) open-loop oscillating, (7) pH process, and (8) processes with large load changes and disturbances.
Since MFA is “Model-Free”, it also overcomes the shortcomings of model-based advanced control methods. MFA is an adaptive and robust control technology but it does not require (1) precise process models, (2) process identification, (3) controller design, and (4) complicated manual tuning of controller parameters. A series of U.S. patents and related international patents for Model-Free Adaptive (MFA) control and optimization technologies have been issued. Some of them are listed in Table 1.
Commercial hardware and software products with Model-Free Adaptive control have been successfully installed in most industries and deployed on a large scale for process control, building control, and equipment control.
Although Model-Free Adaptive (MFA) controllers depart from the traditional control approaches and have solved many difficult control problems, they are mainly used as a component in the traditional process control architecture that comprises a Sensing Layer, Control Layer, Actuation Layer, and Process Layer. There are still many challenging problems in the field of automatic control where traditional process control architecture is no longer sufficient regardless of what controllers are used.
In this patent, we introduce a Self-Organizing Control Architecture that comprises a Sensing Layer, Control Layer, Actuation Layer, Process Layer, as well as Self-Organizing Sensors (SOS), Self-Organizing Actuators (SOA), and Self-Organizing Actuation and Control Units (SOACU). The method and apparatus of SOA and SOACU for process control are presented. A control system as a case example for a gas mixing process is described using the unique SOA and SOACU approaches. A 2x1 Robust MFA controller as a key component of the SOACU is also disclosed.
In the accompanying drawings:
In this patent, the term “mechanism” is used to represent hardware, software, or any combination thereof. The term “process” is used to represent a physical system or process with inputs and outputs that have dynamic relationships. The term “sensor” is used to represent a sensing mechanism. The term “actuator” is used to represent an actuation mechanism or an actuation device in a control system. The term “control loop” refers to a single-loop feedback control system. The term “SISO” refers to Single-Input-Single-Output. The term “2x1” refers to “2-Input-1-Output”. The term “MFA” refers to Model-Free Adaptive control or controllers.
Throughout this document, m=1, 2, 3, . . . , as an integer, which is used to indicate the number of gas or liquid flows in a multi-stream gas or liquid mixing process.
Throughout this document, if a method or apparatus is used to control a gas flow process, it may also be applied to a liquid flow process without departing from the spirit or scope of the invention. If a method or apparatus is used to control a liquid flow process, it may also be applied to a gas flow process without departing from the spirit or scope of the invention.
Throughout this document, if a method or apparatus is related to SOA, it may also be applied to SOACU; and if a method or apparatus is related to SOACU, it may also be applied to SOA without implication of equivalents or departing from the spirit or scope of the invention.
Without losing generality, all numerical values given in this patent are examples. Other values can be used without departing from the spirit or scope of the invention. The description of specific embodiments herein is for demonstration purposes and in no way limits the scope of this disclosure to exclude other non-specifically described embodiments of this invention.
Traditionally, automatic control is based on the concept of feedback. The essence of the feedback theory consists of three components: measurement, comparison, and correction. Measuring the quantity of the variable to be controlled, comparing it with the desired value, and using the error to correct the control action is the basic procedure of feedback automatic control.
r(t)—Setpoint (SP),
PV—Process Variable, PV=x(t)+d(t),
y(t)—Measured Process Variable,
x(t)—Process Output,
u(t)—Controller Output (OP),
d(t)—Disturbance, the disturbance caused by noise or load changes,
e(t)—Error between the Setpoint and Measured Variable, e(t)=r(t)−y(t).
For simplification, the sensor and actuator are typically included as part of the process. Therefore, the Measured Process Variable y(t) can be considered the same as the Process Variable.
The Process Layer includes physical processes or systems with inputs and outputs that have dynamic relationships. For instance, a gas mixing process in an iron and steel plant is a physical process that has multiple process variables to be controlled.
The Sensing Layer includes multiple sensors for measuring various process variables. These sensors can vary significantly in size, type, and physical characteristics. For a gas mixing process, gas flows, gas pressures, and the heating value of mixed gas are measured by their respective sensors.
The Control Layer includes multiple automatic controllers for controlling various process variables. The controllers are typically implemented in control devices such as Distributed Control Systems (DCS), Programmable Logic Controllers (PLC), Programmable Automation Controllers (PAC), Single-Loop Controllers (SLC), or computer software. The controllers include Inputs/Outputs (I/Os), communication buses, or digital networks to interface with sensors and actuators. The Setpoints are the target values for the process variables to track, which are entered, managed, and monitored in the Control Layer. The Control Layer usually includes a Graphical User Interface (GUI) for the operators to monitor the process and control system.
The Actuation Layer includes multiple actuators that take control command signals from the controllers and manipulate certain process inputs or manipulated variables to achieve the control objectives. In a gas mixing process control system, multiple control valves and valve positioners are used as actuators.
A valve positioner is an analog or digital device that controls the valve stem position. It is used to assure that the valve moves to the position that the controller demands. A valve positioner could help deal with variations and issues in packing friction due to dirt, corrosion, lack of lubrication, wear and tear, valve stiction, dead band, and valve nonlinear behavior. It is commonly seen in industrial flow control applications where there are one or more of the following situations: (i) high pressure across the valve, (ii) high pressure applications with tight packing, (iii) valves with wide throttling range, and (iv) valves handling sludge or solids in suspension.
To summarize, a traditional process control architecture may possess the following properties:
1. Multiple sensors for measuring various process variables may exist. However, they send the measurement signals to the Control Layer only;
2. Multiple actuators for controlling different process variables may exist. However, they take commands from the Control Layer only; and
3. A sensor network may exist, but sensors do not talk to each other.
In regard now to the present invention, we will first review the concept of Distributed Intelligence, Self-Organizing, and other related terms in preparation for further discussions of the invention.
Distributed Intelligence can be considered an artificial intelligence method that includes distributed solutions for solving complex problems. It is closely related to Multi-Agent Systems.
Without using strict and academic type definitions, Self-Organizing can be understood as an organization that is achieved in a way that is parallel and distributed. Here, parallel means that all the elements act at the same time, and distributed means no element is a central coordinator.
A self-organizing system is a complex system made up of small and simple units connected to each other and having self-organizing capabilities.
More specifically, the Self-Organizing Process Control Architecture not only comprises the Control Layer 32, Sensing Layer 34, Actuation Layer 36, Process Layer 38, but also one or more of Self-Organizing Sensors (SOS) 40, Self-Organizing Actuators (SOA) 42, and/or Self-Organizing Actuation and Control Units (SOACU) 44.
Notice that the signal flows are not as simple as those of traditional feedback control loops. The Self-Organizing Sensors (SOS), Self-Organizing Actuators (SOA), and Self-Organizing Actuation and Control Units (SOACU) can have direct inputs from the sensor networks. The intelligence has not only been distributed in the sensing, actuation, and control layers, but has also been utilized. The signal flows indicate that this architecture is beyond the scope of traditional control schemes.
This Self-Organizing Process Control Architecture can have one or more of the following properties:
1. Sensors may send measurement signals to other sensors and actuators;
2. A Self-Organizing Actuator (SOA) takes commands from the Controller and may have inputs from sensors;
3. Sensors may talk to each other;
4. A Self-Organizing Sensor (SOS) can have multiple inputs from the sensor networks;
5. A Self-Organizing Sensor (SOS) can send its output to the sensor networks;
6. A Self-Organizing Actuator (SOA) can manipulate multiple manipulated variables in a coordinated way at the same time;
7. A Self-Organizing Actuation and Control Unit (SOACU) incorporates controllers and valve or damper positioners, and provides multiple output signals to manipulate multiple valves, dampers, or other actuation devices in a coordinated way at the same time; and
8. A Multivariable Self-Organizing Actuation and Control Unit (SOACU) can control multiple process variables.
Potential key differences, one or more of which may exist between the traditional process control architecture and the Self-Organizing Process Control Architecture, are compared and summarized in Table 2.
To realize and describe the concept, properties, and significance of the Self-Organizing Process Control Architecture, a realistic actuation scenario is investigated in the context of an industrial process control application where conventional actuators do not work well.
In an iron and steel complex, operating units including blast furnaces, basic oxygen furnaces, and coking ovens all produce gases as byproducts. A gas plant mixes these gases to produce fuel for the furnaces in metal casting and rolling mills. The quality of the mixed gas is measured by its Heating Value. Gases with inconsistent Heating Value can cause safety, product quality, and production problems due to over or under heating.
Even during normal production, gas supply and demand can change randomly and significantly. Major operating units such as blast furnaces in the upstream and reheating furnaces in the downstream may go online and offline periodically causing huge disturbances in gas flows and gas pressures. In order to control the gas mixing process, two control valves are used for each of the gas streams as shown in a 3-stream mixed gas process in
To simplify, we focus on just one gas stream to describe the challenges and methods for controlling a disruptive gas flow process. A disruptive gas flow process is defined to have one or more of the following behaviors: (i) the upstream flow supply and pressure can change significantly; (ii) the downstream flow demand and pressure can change significantly; and/or (iii) the pressure differential between the upstream and downstream gas pipelines is so large that two control valves for each gas flow are required, one for pressure and one for flow.
The objective is to control the gas flow. The pressure controller is required to keep the differential pressure Pd stable so that the gas flow Fg can be effectively controlled. Although this design seems reasonable, it has fundamental flaws. Mainly, these 2 valves are side-by-side trying to control the same gas flow. When the PIC tries to regulate the gas pressure, it affects the gas flow. When the FIC tries to control the gas flow, it affects the gas pressure. So, these two control loops will have a see-saw battle resulting in loop oscillations, inconsistent gas mixing, and large heating value variations. This is a classical industrial process control application, where conventional actuators do not work well.
Since the objective is to control the gas flow, there is no need to have a pressure controller. The system is designed to include a single-loop Flow Controller FIC, and a Self-Organizing Actuator (SOA) that can manipulate the pressure valve and flow valve in a coordinated way at the same time.
The components comprised in the gas flow control system using the traditional approach in
Please note that Pc is the Middle Pressure between the two valves. If an automatic controller is used to control the pressure, only the Middle Pressure can be controlled. The Head Pressure Pa and Back Pressure Pb are dictated by the upstream and downstream processes so that they cannot be controlled. Using the SOA approach, we will not try to control the pressure. The objective is to adjust the pressure to affect the flow valve operating condition. Therefore, the Differential Pressure Pd=Pa−Pb is used as the feedforward signal.
The Self-Organizing Actuator (SOA) can be designed based on the following method:
1. Design the control algorithm and logic so that the pressure control valve can achieve the following objectives: (a) stabilize the differential pressure; (b) regulate the pressure so that the flow control valve works within its relatively linear range such as 25% to 75%, and (c) eliminate or reduce any unnecessary movement of the pressure valve to avoid see-saw battles between the two valves;
2. Incorporate the valve positioning functions into SOA so that external valve positioners are not required;
3. The output (OP) signal of the flow controller may pass through the SOA or may be enhanced by an internal valve positioner to produce output OPf to manipulate the flow valve;
4. The output (OPf) signal of the flow controller is used as a “valve position feedback” signal along with the differential pressure signal Pd for the SOA to produce output OPp to manipulate the pressure valve; and
5. The Robust MFA control technology described in the U.S. Pat. No. 6,684,112 can be incorporated into the design of the Self-Organizing Actuator (SOA).
When incorporating the flow controller FIC with the SOA in
The SOACU 68 comprises an internal flow controller FIC and an internal pressure controller PIC. Designed to work as one unit, the SOACU has two inputs PVp and PVf, two outputs OPp and OPf, a user selectable Setpoint SPf, an internal Setpoint SPu, and an internal feedback PVu. The components and key variables comprised in the gas flow control system using the SOA approach in
Inside the SOACU 68, there is a pressure controller PIC and a flow controller FIC. The PIC has two inputs PVp and PVu, and one output OPp. So, it is a 2-Input-1-Output (2x1) controller. Its setpoint SPu can be set using a pre-determined default value such as 50%, which is the mid point of the “linear” range (25%-75%) of the flow valve. This way, the user does not need to enter a setpoint for the internal PIC controller. The flow controller FIC has one input PVf and one output OPf. Its setpoint SPf is the user selectable target value for the flow.
The control objective is for the Self-Organizing Actuation and Control Unit (SOACU) to produce two outputs OPf and OPp to manipulate the flow valve and pressure valve in a coordinated way so that the gas flow tracks its setpoint SPf under all operating conditions. The SISO MFA controllers that can be used in this embodiment have been described in U.S. Pat. Nos. 6,055,524 and 6,556,980. The 2x1 Robust MFA controller is a unique controller that will be described in
In
The signals shown in
r(t)=SPu—Setpoint of the 2x1 Robust MFA controller,
y(t)=PVu—Process Variable 1 for the 2x1 Robust MFA controller,
u(t)—Primary Controller Output,
e(t)—Error between the Setpoint and Process Variable, e(t)=SPu−PVu,
r1(t)—Upper-bound Controller Setpoint,
r2(t)—Lower-bound Controller Setpoint,
u1(t)—Upper-bound Controller Output,
u2(t)—Lower-bound Controller Output,
ue(t)—The Combined Controller Output,
e1(t)—Error between r1(t) and y(t), e1(t)=r1(t)−y(t),
e2(t)—Error between r2(t) and y(t), e2(t)=r2(t)−y(t),
Pd=PVp—Differential Pressure=Process Variable 2 for the 2x1 Robust MFA controller,
uf(t)—Feedforward MFA Controller Output, and
OPp—2x1 Robust MFA Controller Output.
As shown in
To setup a Robust MFA control system, the user is allowed to enter an Upper-bound (UB) and a Lower-bound (LB) for the Process Variable (PV). These bounds are typically the marginal values that the Process Variable should not go beyond.
It is important to understand that a process variable (PV) is unlike a controller output (OP). A hard limit or constraint can be set for OP since it is a signal produced by a controller. PV is the measured variable for the process output. Its value is a signal obtained from a measurement device such as a sensor. Therefore, trying to limit the PV within a bound can only be done by changing the controller OP to manipulate the process input, which will affect the process output, the PV. To summarize, the PV Upper and Lower bounds are very different than the OP constraints.
The PV Upper and Lower bounds for a Robust MFA controller can be set based on several options as described in the U.S. Pat. No. 6,684,112. In this 2x1 Robust MFA controller case, we can set the bounds relating to the setpoint as follows:
The Upper-bound is based on the primary controller setpoint as follows:
r
1(t)=r(t)+B1, (1)
where B1>0 is a Relative Bound to the setpoint r(t).
The Lower-bound is based on the primary controller setpoint as follows:
r
2(t)=r(t)−B2, (2)
where B2>0 is a Relative Bound to the setpoint r(t).
For instance, if we let B1=B2=25%, a +/−25% upper and lower bound is set around the setpoint r(t). The bounds move as the setpoint changes. For instance, if Setpoint r(t)=50%, Upper-bound=75%, and Lower-bound=25%.
The Constraint Setter 114 is a limit function fc(•) that combines the controller output signals based on the following logic:
u
c(t)=u1(t), if u(t)>u1(t) (3)
u
c(t)=u(t), if u2(t)≦u(t)≦u1(t) (4)
u
c(t)=u2(t), f u(t)<u2(t) (5)
where u1(t) is the output of Upper-bound Controller 100, u2(t) is the output of Lower-bound Controller 102, u(t) is the output of Primary Controller 98, and uc(t) is the output of the limit function fc(•).
SISO MFA controllers can be used for the Primary Controller 98 and the Constraint Controllers 100 and 102. The SISO MFA controllers that can be used in this embodiment have been described in U.S. Pat. Nos. 6,055,524 and 6,556,980. The MFA controller parameters have been described in these patents, which include:
Kc—MFA controller Gain, and
Tc—MFA controller Time Constant.
If the Primary Controller 88 is set with Kc and Tc, the Constraint Controllers 90 and 92 can be set based on, but not limited to, the following formula:
Kc1=α1Kc (6)
Tc1=β1Tc (7)
Kc2=α2Kc (8)
Tc2=β2Tc (9)
where Kc1, Kc2, Tc1, and Tc2, are the MFA Controller Gain and Time Constant for the Upper-bound Controller and Lower-bound Controllers, respectively; and α1, α2, β1, and β2 are positive coefficients that can be set with pre-determined default values or re-configured by the user. For instance, we can let α1=α2=3, and β1=β2=0.7. That means, the Constraint Controllers will have a larger gain and a smaller time constant so that they will react faster compared to the Primary Controller. The objectives of the Constraint Controllers are to limit the PV from going out of pre-determined upper and lower bounds.
As shown in
The Feedforward MFA controllers that can be used in this embodiment have been described in U.S. Pat. Nos. 6,556,980, 6,684,115, and 7,016,743.
The Output Combiner 118 is a function fp(•) that combines the control output signal uc(t) with the Feedforward MFA controller output signal uf(t). It can be designed in different ways. For instance, the output signals can be combined based on the following formula:
OPp=uc(t)+Δuf(t), (10)
where uc(t) is in the range of [0, 100], Δuf(t) is the delta value of uf(t), which is in the range of [−50, 50], and OPp is in the range of [0, 100].
To summarize, the 2x1 Robust MFA controller will provide one or more of the following functions:
1. If there is a big change in differential pressure, the 2x1 controller will take immediate action to regulate the pressure valve to compensate for the change;
2. If the flow valve position is within the pre-determined Upper-bound and Lower-bound, the 2x1 controller will maintain the current differential pressure so that the flow control sub-system can function adequately;
3. If the flow valve position is near or beyond the Upper-bound or Lower-bound, the 2x1 controller will adjust the pressure valve to affect the differential pressure as well as the flow condition so that the flow valve position is gradually moving back within the bound; and
4. If a big disturbance occurs causing the flow valve position to go outside the Upper-bound or Lower-bound quickly, the 2x1 controller will make an immediate control action by adjusting the pressure valve to slow down this momentum. This action will help the flow controller regulate the flow under this abnormal operating condition. In this case, two valves move towards the same direction in a coordinated way at the same time.
In
The trend chart on the top shows Flow SP 130, Flow PV 132, Vf 134, and Vp 136. It can be seen that the SOACU provides good control for the flow where there are large setpoint changes. Please note that after the first setpoint change, Vp 136 goes up quickly in the same direction as Vf 134. During this time, both valves need to open to let the flow increase. Then, while Vf 134 is reduced to keep the flow PV tracking its SP, Vp 136 continues to rise. This action is actually trying to bring Vf down gradually because Vf is higher than the Upper-bound set at 75%. During the second setpoint change, Vp 136 did not go up as quickly as Vf. This is because the pressure valve is already open widely enough to allow the flow to pass through. The self-organizing actions between the two valves are evident by studying the trends.
The trend chart at the bottom shows Pa 142, Pc 144, and Pb 146. The head pressure Pa dictated by the upstream process and back pressure Pb dictated by the downstream process did not change. The middle pressure Pc changed following the pressure valve and flow valve changes as expected.
In
The trend chart on the top shows the PID control loop for the disruptive gas flow with signals of Flow SP 150, PV 152, OP 154. The trend chart in the middle shows the PID control loop for the disruptive gas pressure with signals of Flow SP 156, PV 158, OP 160. As illustrated in the process and instrument diagram of
The trend chart at the bottom shows Pa 162, Pc 164, and Pb 166. The head pressure Pa dictated by the upstream process and back pressure Pb dictated by the downstream process did not change. The middle pressure Pc changes following the pressure valve and flow valve changes.
In
Based on our comprehensive lab simulations and experience in real projects, the control performance of single-loop PID controllers versus Self-Organizing Actuation and Control Unit (SOACU) for controlling a gas-mixing process is analyzed and summarized in Table 5.
Generally speaking, compared with the traditional control approach, the SOA and SOACU approaches demonstrate the following capabilities.
1. Stability of the control system is significantly improved because it avoids the potential conflict of control actions by two control valves;
2. When the upstream or downstream pressure changes, the SOACU shows much smaller disturbance to the flow. This is due to the fact that SOACU has a feedforward MFA controller that takes the differential pressure as a feedforward signal so that it can quickly manipulate the pressure valve to compensate for the pressure disturbances;
3. The PID controllers are sensitive to tuning parameters. It is seen that the flow and pressure loops may show oscillations when working in nonlinear range;
4. The PID based pressure and flow control loops may fight each other resulting in undesirable control performance. The SOACU can manipulate the pressure valve and flow valve in a coordinated way. It can be seen that when there is a setpoint change, the flow control output OPf changes quickly to perform flow control. On the other hand, the pressure control output may or may not change depending on the position of the flow output; and
5. In SOACU, the pressure control output OPp may move gradually to affect the flow valve position so it moves back within the upper bound set at 75%. This demonstrates that its control outputs OPf and OPp act in a coordinated way.
The features and benefits of an SOA and SOACU based control system include:
1. The Self-Organizing Actuation and Control Units (SOACU) are developed based on a general-purpose approach where Model-Free Adaptive (MFA) controllers are used. The solution can effectively deal with large and random flow and pressure disturbances due to sudden changes in the flow supply and demand from the upstream and downstream processes;
2. Using the SOACU technology, the flow in each stream can be effectively controlled, and the differential pressure can be stabilized during disruptive operating conditions. Therefore, the interactions among the flow streams may still exist but are significantly reduced;
3. Since the SOACU allows the flow valves to work in their relatively linear range, valve positioners for the flow valves may not be required. This will result in cost savings;
4. Internal valve positioners can still be designed as part of the SOACU to deal with more difficult control and actuation situations; and
5. Since SOACU incorporates all the instrumentation and actuation devices, this complex system becomes much more concise and easier to implement and maintain.
This application claims priority to U.S. Provisional Application No. 61/812,143 filed on Apr. 15, 2013, which is herein incorporated by reference.
This invention was made with government support under SBIR grant DE-SC0008235 and SBIR grant DE-FG02-08ER84944 awarded by the U.S. Department of Energy. The government has certain rights to the invention.
| Number | Date | Country | |
|---|---|---|---|
| 61812143 | Apr 2013 | US |
