Control loop monitoring and diagnosis method and related device

Abstract

The invention discloses a control loop monitoring and diagnosis method and a related device. Real-time monitoring and accurate diagnosis of an industrial system control loop are achieved. The method comprises the following steps: receiving a data monitoring request for a specified control loop; traversing a connecting line relation list corresponding to the specified control loop, and searching any connecting line relation containing a target element which is the name of a front pin or a rear pin of any connecting line connected with a predetermined key node in the specified control loop; determining the ID of the key node according to the searched connection relationship and the inherent corresponding relationship between the key node and the connection line thereof; traversing a node information list corresponding to the specified control loop, searching node information containing the ID of the key node, extracting operation monitoring data of the node from the searched node information, and taking the operation monitoring data as input data of a pre-constructed equipment diagnosis model; and the equipment diagnosis model calculates to obtain an operation condition diagnosis result of the specified control loop.

Description

Technical Field

[0001] The present application relates to the technical field of industrial intelligent monitoring and equipment diagnosis, and in particular to a control loop monitoring and diagnosis method and related devices. Background Art

[0002] Industrial systems are the core support architecture for the operation of process industries. However, the means of monitoring the operating conditions of their control loops (there are multiple control loops in industrial systems, such as flow control loops, temperature control loops, liquid level control loops, and cascade control loops, etc.) are relatively weak under existing technical conditions, resulting in many potential safety hazards being difficult to effectively identify and eliminate. Once a safety accident occurs, it usually leads to extremely serious consequences. Summary of the invention

[0003] In view of the above problems, the present application provides a control loop monitoring and diagnosis method and related devices to achieve real-time monitoring and accurate diagnosis of industrial system control loops. The specific scheme is as follows:

[0004] A first aspect of the present application provides a control loop monitoring and diagnosis method, comprising:

[0005] Receive a data monitoring request for a specified control loop; wherein there are multiple control loops in the industrial system, each control loop includes multiple nodes, each node is equipped with at least one pin, and the pin of the node is connected to the corresponding pin of other nodes in the same control loop through a wire; a wire relationship list and a node information list are stored in the database of the industrial system for each control loop; the wire relationship recorded in the wire relationship list for each wire in the control loop includes the name of the leading pin of the wire, the ID of the leading node, the name of the trailing pin and the ID of the trailing node; the node information recorded in the node information list for each node in the control loop includes the ID and operation monitoring data of the node;

[0006] Traversing the connection relationship list corresponding to the specified control loop, searching for any connection relationship containing a target element, wherein the target element is the name of a leading or trailing pin of any connection connected to a predetermined key node in the specified control loop; then, determining the ID of the key node according to the found connection relationship and the inherent correspondence between the key node and its connection;

[0007] Traversing the node information list corresponding to the specified control loop, searching for the node information containing the ID of the key node, and extracting the operation monitoring data of the node from the found node information;

[0008] The extracted operation monitoring data is used as input data of a pre-built equipment diagnosis model, and the equipment diagnosis model performs calculations based on the input data to obtain an operation condition diagnosis result of a specified control loop.

[0009] In a possible implementation, the operation monitoring data of the node includes the real-time value and historical value of each pin of the node, the real-time value of the operation parameter of the node, and the real-time value of the monitoring parameter of the node.

[0010] In a possible implementation, the operating condition diagnosis result is reflected in three dimensions, namely: control performance of the control loop, instrument health status of the control loop, and oscillation and nonlinear status of the control loop;

[0011] The control performance of the control loop refers to a comprehensive evaluation of the control effect of the control loop;

[0012] The instrument health status of the control loop refers to a comprehensive evaluation of the health status of each instrument in the control loop;

[0013] The oscillation and nonlinear state of the control loop refers to a comprehensive evaluation of abnormal phenomena induced by oscillation or nonlinear factors in the control loop.

[0014] In a possible implementation, the control performance of the control loop is quantified by multiple indicators, including: a stable segment control performance indicator, a sluggishness indicator, an MV stability indicator, a PV-SV deviation, an environmental interference degree, and a MV overlimit percentage;

[0015] The stable segment control performance index is used to evaluate the fluctuation of the process variable PV around the set value SV;

[0016] The sluggishness index is used to measure the hysteresis degree of the control loop response;

[0017] The MV stability index is used to evaluate the stability of the manipulated variable MV in the control loop;

[0018] The PV-SV deviation refers to the degree of deviation between the process variable PV and the set value SV;

[0019] The environmental interference degree is used to measure the influence of external environmental factors on the control loop;

[0020] The MV limit exceeding percentage refers to the proportion of the manipulated variable MV exceeding its allowable range;

[0021] Each indicator under the control performance of the control loop divides the state of the control loop into three categories: fault, abnormal and normal: when the indicator value exceeds the preset acceptable range, the control loop is judged to be in a fault state; if the indicator value has not reached the fault threshold but has exceeded the normal range, the control loop is judged to be in an abnormal state; and when the indicator value is within the normal range, it indicates that the control loop is in a normal state.

[0022] In a possible implementation, the instrument health status of the control loop is quantified by multiple indicators, including: valve viscosity comprehensive index, valve stem sliding index, valve stem dead zone index, MV-PV viscosity mode index index and sensor accuracy index;

[0023] The valve viscosity comprehensive index is used to evaluate the viscosity of the valve during movement;

[0024] The valve stem sliding index is used to quantify the frequency and amplitude of valve stem sliding;

[0025] The valve stem dead zone index is used to quantify the range of the valve stem movement without causing significant changes in the process variable;

[0026] The MV-PV sticky mode index is used to evaluate the stickiness between the manipulated variable MV and the process variable PV;

[0027] The sensor accuracy index is used to evaluate the measurement accuracy and stability of the sensor;

[0028] Each indicator of the instrument health status of the control loop divides the status of the control loop into three categories: fault, abnormal and normal: when the indicator value exceeds the preset acceptable range, the control loop is judged to be in a fault state; if the indicator value has not reached the fault threshold but has exceeded the normal range, the control loop is judged to be in an abnormal state; and when the indicator value is within the normal range, it indicates that the control loop is in a normal state.

[0029] In a possible implementation, the oscillation and nonlinear state of the control loop are quantified by multiple indicators, including: an oscillation indicator, an oscillation period indicator, an oscillation spectrum power percentage indicator, and a nonlinearity indicator;

[0030] The oscillation index is used to quantify the oscillation phenomenon that occurs during the operation of the control loop;

[0031] The oscillation period indicator is used to control the time interval of repeated oscillation phenomena in the loop;

[0032] The oscillation spectrum power percentage index is used to quantify the relative strength of different frequency oscillation components in the control loop;

[0033] The nonlinear index is used to quantify the influence of nonlinear characteristics in the control loop;

[0034] Each indicator under the oscillation and nonlinear state of the control loop divides the state of the control loop into three categories: fault, abnormal and normal: when the indicator value exceeds the preset acceptable range, the control loop is judged to be in a fault state; if the indicator value has not reached the fault threshold but has exceeded the normal range, the control loop is judged to be in an abnormal state; and when the indicator value is within the normal range, it indicates that the control loop is in a normal state.

[0035] In a possible implementation, each indicator is displayed via a key performance indicator KPI historical baseline chart.

[0036] In a possible implementation, the connection relationship set for each connection in the control loop in the connection relationship list further includes: the name of the preceding node and the name of the succeeding node of the node;

[0037] The node information list records the node information of each node in the control loop and further includes: the name of the node;

[0038] The connection relationship list further includes the connection relationship set for each connection in the control loop: the ID of the front pin and the ID of the rear pin of the node;

[0039] The database of the industrial system also stores, for each control loop, the following information: the ID of the control loop, the type of the control loop, the name of the control loop, and the description of the control loop.

[0040] In a possible implementation, after extracting the operation monitoring data of the node from the found node information, the method further includes: storing the node type as a key node in the found node information;

[0041] The control loop monitoring and diagnosis method also includes: when a data monitoring request for the same designated control loop is received again, traversing the node information list, searching for node information containing a node type of a key node, and extracting the node's operation monitoring data from the found node information; using the extracted operation monitoring data as input data of a pre-built equipment diagnosis model, the equipment diagnosis model performs calculations based on its input data to obtain an operating condition diagnosis result of the designated control loop.

[0042] A second aspect of the present application provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements the control loop monitoring and diagnosis method of the first aspect or any implementation method of the first aspect.

[0043] Through the above technical solutions, the control loop monitoring and diagnosis method provided by this application is based on the technology of big data analysis and artificial intelligence, and realizes the innovation of control loop monitoring and diagnosis for industrial systems with complex influencing factors, a large number of loops, cumbersome management, and delayed performance degradation processing. It can monitor the operating status of any control loop in the industrial system in real time, and accurately obtain the operation monitoring data of key nodes by traversing the connection relationship list and the node information list. These data provide strong support for the dynamic perception of the risk of the control loop, and realize the step-by-step fault location mode from surface to line and then to point. This method can not only help users to efficiently perform fault warning and predictive diagnosis, significantly improve the efficiency of operation and maintenance work, but also guide users to implement targeted preventive maintenance, effectively avoiding the losses caused by subsequent maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0045] Figure 1 A flow chart of a control loop monitoring and diagnosis method provided in this application;

[0046] Figure 2 A schematic diagram of a single-loop structure provided for this application; DETAILED DESCRIPTION

[0047] Process industry, also known as process industry or continuous industry, is a type of industry that uses a series of processing devices to make raw materials undergo specified chemical reactions or physical changes to obtain the desired products. The production process of this type of industry is usually continuous and needs to maintain a stable operating state to ensure product quality and production efficiency.

[0048] With the advancement of technology and market demand, process industries such as chemical, petrochemical, pharmaceutical, and food processing are experiencing rapid development. This development has brought about a rapid increase in the number and types of equipment, thereby improving production efficiency, but it has also brought about a series of challenges in operation and maintenance management, such as increased operation and maintenance workload and increased safety risks. What is more serious is that the industrial system, as the core support architecture for the operation of the process industry, has relatively weak monitoring methods for its control loop (there are multiple control loops in the industrial system, such as flow control loops, temperature control loops, liquid level control loops, and cascade control loops; a controller receives an input quantity and calculates and determines an output quantity according to certain control rules and algorithms, so that a control loop is formed between the input and the output; each control loop contains multiple nodes, each node is equipped with at least one pin, and the pin of the node is connected to the corresponding pins of other nodes in the same control loop through a wire, thereby establishing a connection relationship between the nodes) operating conditions, which makes it difficult to effectively identify and eliminate many potential safety hazards. Once a safety accident occurs, it usually leads to extremely serious consequences and directly causes huge economic losses.

[0049] Specifically, under existing technical conditions, the monitoring methods for the operating conditions of control loops in industrial systems mainly have the following significant problems:

[0050] 1) Complex influencing factors: The performance of the control loop is affected by a variety of factors, which are intertwined, making it extremely difficult to accurately determine the root cause of the fault. This places extremely high demands on the professional skills and diagnostic capabilities of operation and maintenance personnel.

[0051] 2) A large number of circuits and complicated management: Faced with a large number of control circuits, it is difficult to quickly and accurately locate the problem circuit. In addition, the inspection and maintenance process is complicated and time-consuming, which not only increases the workload of operation and maintenance personnel, but also leads to high personnel investment costs.

[0052] 3) Delayed processing of performance degradation: For unexplained degradation of control performance, it is often difficult for instrumentation personnel to detect and handle it in a timely manner. This not only poses the risk of omission and misjudgment, but may also affect production efficiency and product quality.

[0053] In this regard, an embodiment of the present application provides a control loop monitoring and diagnosis method. This method is based on the technology of big data analysis and artificial intelligence. In the face of industrial systems with complex influencing factors, a large number of loops, cumbersome management, and delayed performance degradation processing, it can monitor the operating status of any control loop in the industrial system in real time, obtain rich operating data and fault data, and provide strong support for the dynamic perception of risks in the control loop. This can not only help users to perform fault warning and predictive diagnosis on key equipment, thereby significantly improving the efficiency of operation and maintenance work, but more importantly, it can also guide users to implement targeted preventive maintenance, effectively avoid the losses caused by subsequent maintenance, thereby fundamentally improving the overall quality of operation and maintenance work.

[0054] A control loop monitoring and diagnosis method provided by an embodiment of the present application is described in detail below in conjunction with the accompanying drawings. A person skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solution provided by the embodiment of the present application is also applicable to similar technical problems.

[0055] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0056] A control loop monitoring and diagnosis method provided by an embodiment of the present application is applied to an industrial system. A connection relationship list and a node information list are stored in the database of the industrial system for each control loop; the connection relationship list records a connection relationship for each connection in the control loop, and the connection relationship of each connection at least includes four items of information: the name of the leading pin of the connection, the ID of the leading node of the connection, the name of the trailing pin of the connection, and the ID of the trailing node of the connection; the node information list records a node information for each node in the control loop, and the node information of each node at least includes two items of information: the ID of the node and the operation monitoring data of the node.

[0057] See also Figure 1 , a control loop monitoring and diagnosis method provided by an embodiment of the present application specifically includes:

[0058] Step S01: receiving a data monitoring request for a designated control loop, and then proceeding to step S02.

[0059] Specifically, the user may submit a request for data monitoring of one or more control loops (specified by the user) to the controller of the industrial system at the same time.

[0060] Before monitoring begins, pre-select one of the most representative nodes in the control loop as a key node. A key node refers to a node that plays a vital role or has unique functions in the control loop, and changes in its state or performance have a significant and direct impact on the operating efficiency and stability of the entire control loop. In order to effectively monitor the operating conditions of the control loop, this can be achieved by continuously collecting and analyzing the operating monitoring data of the key node in the recent period (for example, the past few hours, days, or a time period set according to actual needs) (the operating monitoring data of the node generally includes the real-time and historical values ​​of each pin of the node, the real-time values ​​of the operating parameters of the node, and the real-time values ​​of the monitoring parameters of the node). This monitoring method can not only help users promptly discover possible problems or anomalies in the control loop, but also provide valuable data support for subsequent troubleshooting, performance optimization, and system maintenance.

[0061] Step S02: traverse the list of connection relationships corresponding to the specified control loop, and search for any connection relationship that contains a target element, where the target element is the name of the leading or trailing pin of any connection connected to a predetermined key node in the specified control loop (the names of the leading and trailing pins of the connection connected to the key node are known, but it is unknown which node in the control loop is the key node); then, determine the ID of the key node based on the connection relationship found and the inherent correspondence between the key node and its connection; then proceed to step S03.

[0062] Specifically, a control loop consists of one or more function blocks and bit numbers, and its concepts are explained as follows:

[0063] 1. Functional Block

[0064] Function blocks are the basic building blocks of control programs in control loops, and are used to implement specific control logic or data processing functions. These function blocks come in many different types, each designed to meet specific control requirements. Users can select and configure these function blocks according to actual needs to build a complete control loop control program.

[0065] 2. Position number

[0066] Instrument hardware (including data sources and output sources) is connected to the controller through physical connections (such as cables, wires, etc.). The data source is responsible for collecting various parameter information and transmitting it to the controller, while the output source is responsible for performing corresponding actions according to the controller's instructions. In the control loop, each data source or output source is assigned a unique identifier, namely a bit number. The bit number is used to identify different data sources or output sources.

[0067] Function blocks and bit numbers are nodes in the control loop. Nodes are connected to each other through their respective output pins and input pins. Pins are elements responsible for connection in function blocks or bit numbers, representing channels for data to flow in or out. The connection between the output pin and the input pin is regarded as an object. The embodiment of the present application defines that each connection object has four key attributes: the ID of the preceding node (i.e., the node that inputs data to the connection), the ID of the following node (i.e., the node that receives the data transmitted by the connection), the name of the preceding pin (i.e., the pin in the preceding node that transmits data to the connection) and the name of the following pin (i.e., the pin in the following node that receives the data transmitted by the connection).

[0068] The types of nodes in the control loop include not only key nodes, but also AO nodes, main loop nodes, main loop AO nodes, main loop AI nodes, main loop other nodes, sub-loop nodes (the main loop is the main data control loop, which usually determines the input of the sub-loop and must be upstream of the sub-loop; the sub-loop is the sub-control loop, which is usually used to directly control the output value of the bit number and must be downstream of the main loop), sub-loop AO nodes, sub-loop AI nodes, sub-loop other nodes, DO nodes, DI nodes, input other nodes, output other nodes, other nodes, etc.

[0069] There is an inherent correspondence between a specific type of node (such as a key node) and its connection. Then, according to the properties of all connection objects and the inherent correspondence between a specific type of node and its connection, the ID of the specific type of node can be determined.

[0070] Take a single loop (also called a simple control loop) as an example. Figure 2As shown, the single loop includes three nodes, and the types of these three nodes are other nodes (left nodes), key nodes (represented by PIDEX) and AO nodes (right nodes). The output pin BKOUT of the other node is connected to the input pin BKIN of the key node through a wire, and another output pin BKOUT of the other node is connected to the input pin BKINERR of the key node through a wire, and the output pin PV of the other node is connected to the input pin PV of the key node through a wire, and the output pin ERR of the other node is connected to the input pin PVERR of the key node through a wire, and the output pin MV of the key node is connected to the input pin IN of the AO node through a wire. It is known that Figure 2 By using the four key attributes of each connection in the circuit, we can determine which node in the circuit is the key node and what the ID of the key node is. For example, if the name of the leading pin of a certain connection is known to be "PV", then according to the inherent correspondence between the key node and its connection, we can determine that the trailing node of the connection is the key node in the single circuit, and then combined with the key attributes of the connection, we can determine the ID of its trailing node; or, if the name of the leading pin of a certain connection is known to be "MV", then according to the inherent correspondence between the key node and its connection, we can determine that the leading node of the connection is the key node in the single circuit, and then combined with the key attributes of the connection, we can determine the ID of its leading node.

[0071] Therefore, in order to facilitate the determination of the key node and the ID of the key node among the many nodes of a control loop, a connection relationship list and a node information list can be set in advance for the control loop. The connection relationship recorded in the connection relationship list for each connection in the control loop includes the name of the front pin of the connection, the ID of the front node, the name of the rear pin and the ID of the rear node; the node information recorded in the node information list for each node in the control loop includes the ID and operation monitoring data of the node (the operation monitoring data of the node generally includes the real-time value and historical value of each pin of the node, the real-time value of the operating parameter of the node and the real-time value of the monitoring parameter of the node). Then, when a data monitoring request for a control loop is received, by traversing the connection relationship list corresponding to the control loop, any connection relationship containing the target element (the target element is the name of the front or rear pin of any connection connected to the key node of the control loop) is searched; then, according to the connection relationship found and the inherent correspondence between the key node and its connection, the ID of the key node can be determined.

[0072] When obtaining the historical value of a pin, the user can select the time range according to their needs without limitation. Figure 2For example, if the selected time range is 4 hours, and data from one time point is taken every 1 second, each pin will have a maximum of 14,400 pieces of data information; if the selected time range is 1 day, and data from a time point is taken every 10 seconds, each pin will have a maximum of 8,640 pieces of data information; if the selected time range is 7 days, and data from a time point is taken every 5 minutes, each pin will have a maximum of 2,016 pieces of data information; if the selected time range is 30 days, and data from a time point is taken every 10 minutes, each pin will have a maximum of 4,320 pieces of data information.

[0073] In addition, to increase the readability of the information, the connection relationship list for each connection in the control loop may also include: the name of the preceding node of the node, the name of the following node of the node. The node information list for each node in the control loop may also include: the name of the node. In addition, to facilitate the distinction between different pins with the same name, the connection relationship list for each connection in the control loop may also include: the ID of the preceding pin of the node and the ID of the following pin of the node. In addition, the database of the industrial system also stores for each control loop: the ID of the control loop, the type of the control loop (such as temperature control, pressure control, etc.), the name of the control loop, and the description of the control loop (including its function, purpose, etc.).

[0074] Step S03: traverse the node information list corresponding to the specified control loop, search for the node information containing the ID of the key node, and extract the operation monitoring data of the node from the found node information; then enter step S04.

[0075] Specifically, when the user terminal extracts the operation monitoring data of the node, the communication method between the user terminal and the controller is, for example, WebSocket. WebSocket is a protocol for full-duplex communication on a single TCP (Transmission Control Protocol) connection, which allows the controller to actively push data to the user terminal without the user terminal polling the controller. This enables the system to update the operation monitoring data in real time and provide instant feedback to the user. The user can connect to the controller through WebSocket and subscribe to the operation monitoring data of the PIDEX function block of the key node. When the operation monitoring data changes, the controller will actively push the latest operation monitoring data to the user terminal. The user terminal can display these values ​​in real time and perform corresponding control operations as needed.

[0076] Step S04: The extracted operation monitoring data is used as input data of a pre-built equipment diagnosis model. The equipment diagnosis model performs calculations based on the input data to obtain an operation condition diagnosis result of a specified control loop.

[0077] Specifically, the operation monitoring data of key nodes generally include the real-time value and historical value of each pin of the node, the real-time value of the operation parameters of the node, and the real-time value of the monitoring parameters of the node. Among them, each pin generally includes: PV (closed-loop control process measurement value) pin, CSV (cascade input / external setting) pin, TV (tracking input value) pin, MF (PID actuator feedback value, used for monitoring) pin, MV (PID adjustment output value) pin and SV (closed-loop control set value) pin. The operation parameters of the node generally include: PB (proportional size) parameter, TI (integral time) parameter, TD (differential time) parameter, KD (differential filter coefficient) parameter. The monitoring parameters of the node generally include: MODE (working mode) parameter. The operation monitoring data of the key nodes of the control loop reflects the quality of the operating conditions of the control loop and provides an important reference for evaluating the operating conditions of the control loop.

[0078] In this control loop monitoring and diagnosis method, a large number of connection relationship lists and node information lists stored in the industrial system database are traversed to find and extract the operation monitoring data of key nodes in the specified control loop. This process fully demonstrates the powerful ability of big data analysis technology in processing and analyzing massive data, and can efficiently screen out key information to provide strong support for subsequent analysis and diagnosis. The use of pre-built equipment diagnosis models to intelligently calculate and analyze the extracted operation monitoring data to obtain the diagnosis results of the operating conditions of the control loop reflects the excellent capabilities of artificial intelligence in data processing, pattern recognition and predictive analysis.

[0079] In summary, the control loop monitoring and diagnosis method provided by the embodiment of the present application is based on the technology of big data analysis and artificial intelligence, and realizes the innovation of control loop monitoring and diagnosis for industrial systems with complex influencing factors, large number of loops, cumbersome management, and delayed performance degradation processing. It can monitor the operating status of any control loop in the industrial system in real time, and accurately obtain the operation monitoring data of key nodes by traversing the connection relationship list and the node information list. These data provide strong support for the dynamic perception of the risk of the control loop, and realize the step-by-step fault location mode from surface to line and then to point. This method can not only help users to efficiently perform fault warning and predictive diagnosis, significantly improve the efficiency of operation and maintenance work, but also guide users to implement targeted preventive maintenance, effectively avoiding the losses caused by subsequent maintenance. This method makes deep use of data, optimizes intelligent management of equipment, reduces the workload of equipment management personnel, improves the maintenance quality of equipment and the quality of overall operation and maintenance work, and fundamentally guarantees the stable operation and efficient output of industrial systems.

[0080] Among them, the operating condition diagnosis of the control loop mainly focuses on three key dimensions: the first is the control performance of the control loop, the second is the instrument health status of the control loop, and the third is the oscillation and nonlinear state of the control loop. Each dimension can be quantified through a series of specific indicators. In addition, each indicator divides the state of the control loop into three categories: fault, abnormal, and normal: when the indicator value exceeds the preset acceptable range, the control loop is considered to be in a fault state, indicating that there is a serious problem that needs to be dealt with immediately; if the indicator value does not reach the fault threshold but exceeds the normal range, it is judged to be an abnormal state, indicating that there may be potential problems or further optimization; and when the indicator value is within the normal range, it indicates that the control loop is running well and is in a normal state. Therefore, when at least one indicator value exceeds the preset acceptable range, the control loop is diagnosed as a fault state; when at least one indicator value does not reach the fault threshold but exceeds the normal range and all other indicators are within the normal range, the control loop state is diagnosed as normal; when all indicator values ​​are within the normal range, the control loop state is diagnosed as normal.

[0081] These three dimensions are described in detail below:

[0082] 1. Control performance

[0083] Control performance evaluation refers to a comprehensive evaluation of the control effect of the control loop, which mainly examines whether the control loop responds to changes in the set value stably, accurately and quickly, and whether there are excessive deviations or fluctuations.

[0084] The control performance evaluation results can be quantified through a series of specific control performance indicators, including: stable segment control performance index, slowness index, MV stability index, PV-SV deviation, environmental interference and MV over-limit percentage, etc. The specific analysis of each indicator is as follows:

[0085] 1) Steady-state control performance indicators

[0086] The plateau control performance index is used to evaluate the fluctuation of the process variable (PV) around the set point (SV).

[0087] For example, the stable segment control performance index value is set between 0 and 1. The smaller the value, the worse the control effect. A value less than or equal to 0.1 is a fault, a value greater than 0.1 and less than 0.4 is an abnormality, and the rest are normal.

[0088] 2) Slow indicators

[0089] The sluggishness index is used to measure the lag in the control loop response. It reflects the time delay between the control loop receiving the input signal and starting to respond.

[0090] For example, the value of the sluggishness index can be set between 0 and 1. A larger value represents a worse control effect. A value greater than or equal to 0.75 represents a fault, a value greater than 0.5 and less than 0.75 represents an abnormality, and the rest are normal.

[0091] 3) MV stability index

[0092] The MV stability index is used to evaluate the stability of the manipulated variable (MV) in a control loop. It is usually quantified by calculating the fluctuation range or standard deviation of the MV over a certain period of time.

[0093] For example, the value of the MV stability index can be set between 0 and 1. A smaller value means a worse control effect. A value less than or equal to 0.1 means a fault, a value greater than 0.1 and less than 0.3 means an abnormality, and the rest are normal.

[0094] 4) PV-SV deviation

[0095] PV-SV deviation refers to the degree of deviation between the process variable (PV) and the set value (SV). It reflects the accuracy of the control loop in achieving the set target.

[0096] For example, the PV-SV deviation scale value can be set between 0-100. The larger the value, the worse the control effect. A value greater than or equal to 5 indicates a fault, a value greater than 2 and less than 5 indicates an abnormality, and the rest are normal.

[0097] 5) Environmental interference

[0098] Environmental interference is used to measure the impact of external environmental factors on the control loop. It includes physical factors such as noise, vibration, electromagnetic interference, and environmental factors such as temperature and humidity that interfere with the performance of the control loop.

[0099] For example, the value of the environmental interference degree can be set between 0-100. The larger the value, the worse the control effect. A value greater than or equal to 5 indicates a fault, a value greater than 2 and less than 5 indicates an abnormality, and the rest are normal.

[0100] 6) MV over limit percentage

[0101] MV overlimit percentage refers to the proportion of the manipulated variable (MV) exceeding its allowable range. It reflects whether the manipulated variable remains within a safe and effective range during operation of the control loop.

[0102] For example, the MV over-limit percentage value can be set between 0-100. The larger the value, the worse the effect. A value greater than or equal to 50 indicates a fault, a value greater than 20 and less than 50 indicates an abnormality, and the rest are normal.

[0103] 2. Instrument health status

[0104] The instrument health index refers to a comprehensive evaluation of the health status of each instrument in the loop, including whether the control valve is sticky or the sensor is faulty.

[0105] The instrument health assessment results can be quantified through a series of specific instrument health indicators, including: valve viscosity comprehensive index, valve stem slip index, valve stem dead zone index, MV-PV viscosity mode index index and sensor accuracy index. The specific analysis of each indicator is as follows:

[0106] 1) Comprehensive index of valve viscosity:

[0107] The valve sticking index is used to evaluate the sticking degree of the valve during movement. Sticking refers to the unsmooth movement of the valve during opening or closing due to friction, dirt or other factors. The valve sticking index quantifies the sticking degree of the valve by comprehensively considering the valve's motion characteristics, response time and other factors.

[0108] For example, the value of the valve viscosity comprehensive index can be set between 0-1. The larger the value, the worse the effect. A value greater than or equal to 0.6 indicates a fault, a value greater than 0.45 and less than 0.6 indicates an abnormality, and the rest are normal.

[0109] The valve sticking diagnosis module includes a high-order statistical nonlinear diagnosis algorithm, a valve XCH model system identification algorithm and a MV-PV mode diagnosis algorithm.

[0110] Among them, the high-order statistical nonlinear diagnostic algorithm uses high-order statistics (such as skewness, kurtosis, etc.) to detect nonlinear characteristics in the system, including viscosity. By analyzing the statistical characteristics of the system response signal, the algorithm can identify nonlinear factors such as viscosity, providing a basis for subsequent fault diagnosis and repair.

[0111] The valve XCH model system identification algorithm identifies the dynamic characteristics of the valve based on the valve's XCH (cross-coupling) model; by constructing a mathematical model of the valve, the algorithm can accurately describe the valve's motion characteristics and response laws, thereby achieving effective diagnosis of valve faults such as valve sticking.

[0112] The MV-PV mode diagnosis algorithm detects whether there are faults such as sticking in the system by analyzing the change pattern between the manipulated variable (MV) and the process variable (PV); the algorithm can identify the abnormal change relationship between MV and PV, thereby realizing timely detection and early warning of control system faults.

[0113] 2) Valve stem sliding index:

[0114] Stem slippage refers to the phenomenon that the valve stem briefly pauses or jumps during the opening or closing process due to friction, vibration or unstable control signals. The stem slippage index is used to quantify the frequency and amplitude of the stem slippage to evaluate the stability and reliability of the valve.

[0115] For example, the valve stem slip index value can be set between 0-20. The larger the value, the worse the effect. A value greater than or equal to 4 indicates a fault, a value greater than 1 and less than 4 indicates an abnormality, and the rest are normal.

[0116] 3) Valve stem dead zone index:

[0117] Stem deadband refers to the movement of the valve stem within a certain range without causing a significant change in the process variable (PV). This is usually caused by valve friction, sealing performance or insufficient control signal accuracy. The valve stem deadband indicator is used to quantify the size of this range to evaluate the sensitivity and control accuracy of the valve.

[0118] For example, the value of the valve stem dead zone indicator can be set between 0-20. The larger the value, the worse the effect. A value greater than or equal to 4 indicates a fault, a value greater than 1 and less than 4 indicates an abnormality, and the rest are normal.

[0119] 4) MV-PV viscosity mode index:

[0120] The MV-PV sticky mode index is used to evaluate the stickiness between the manipulated variable (MV) and the process variable (PV). This index quantifies the stickiness between the MV and PV by analyzing the response time, rate of change and other parameters of the two to evaluate the dynamic performance and stability of the control loop.

[0121] For example, the value of the MV-PV viscosity mode index indicator can be set between 0 and 1. The larger the value, the worse the effect. A value greater than or equal to 0.3 indicates a fault, a value greater than 0.2 and less than 0.3 indicates an abnormality, and the rest are normal.

[0122] 5) Sensor accuracy indicators:

[0123] The sensor accuracy index is used to evaluate the measurement accuracy and stability of the sensor. It reflects the degree of deviation between the output signal and the true value of the sensor during the measurement process.

[0124] For example, the value of the sensor accuracy index can be set between 0 and 1. The smaller the value, the worse the effect. A value less than or equal to 0.25 indicates a fault, a value greater than 0.25 and less than 0.6 indicates an abnormality, and the rest are normal.

[0125] 3. Oscillation and nonlinear state

[0126] The oscillation and nonlinearity of the control loop can be quantified through a series of specific instrument health indicators, including: oscillation index, oscillation period index, oscillation spectrum power percentage index and nonlinearity index. The specific analysis of each indicator is as follows:

[0127] 1) Oscillator

[0128] Oscillation indicators are used to quantify the oscillations that occur during the operation of a control loop.

[0129] 2) Oscillation cycle indicator

[0130] The oscillation period indicator describes the time interval at which oscillations in a control loop recur.

[0131] 3) Oscillation spectrum power percentage index

[0132] The oscillation spectrum power percentage indicator is used to quantify the relative strength of oscillation components of different frequencies in the control loop.

[0133] 4) Nonlinear indicators

[0134] Nonlinearity indicators are used to quantify the impact of nonlinear characteristics in a control loop.

[0135] If the oscillation index, oscillation period index, oscillation spectrum power percentage index and nonlinear index are all within the normal range, the control loop is running stably and no additional measures are required. At this time, the control loop status is diagnosed as normal. For oscillations caused by mild nonlinear factors or other minor reasons, the control loop performance may decline, but it can still maintain its basic functions. In this case, the control loop status is diagnosed as abnormal. Although the function is not completely lost, it has deviated from the optimal state. For oscillations caused by severe nonlinear factors or other major reasons, the control loop performance will be significantly reduced, which may lead to functional failure or even damage to the control loop. At this time, the control loop status should be diagnosed as a fault, and immediate measures need to be taken to repair or replace it.

[0136] In a possible implementation, in order to facilitate users to quickly and accurately observe the indicator change trend of the control loop within a specific time range, a KPI (Key Performance Indicator) historical baseline graph of the control loop can be drawn. In this graph, the horizontal axis represents time and the vertical axis corresponds to the corresponding KPI indicator value. That is, each indicator is displayed through the KPI historical baseline graph.

[0137] The KPI historical baseline chart is a chart used to show how key performance indicators (KPIs) change over time. It uses specific historical data as a baseline reference. By organizing and plotting KPI data from different stages in the past, the development trajectory of key indicators can be intuitively presented, helping managers and other relevant personnel to clearly understand the performance of the business over the past period of time, and then analyze the changing trends and fluctuation ranges of indicators, providing a strong visual basis for comparing current performance, setting reasonable goals, and formulating subsequent improvement strategies.

[0138] In a possible implementation, after extracting the operation monitoring data of the node from the found node information, the method further includes: storing the node type as a key node in the found node information.

[0139] When a data monitoring request for the same specified control loop is received again, the node information list is traversed to find node information containing key nodes, and the operation monitoring data of the node is extracted from the found node information; the extracted operation monitoring data is used as input data of the pre-built equipment diagnosis model, and the equipment diagnosis model is calculated based on its input data to obtain the operation condition diagnosis result of the specified control loop. Therefore, for any specified control loop, it is only necessary to pass the first time a data monitoring request is received for it. Figure 1 The solution shown is to search and obtain the operation monitoring data, and then you can search and obtain it directly by node type, which is more convenient and quick.

[0140] In addition, an embodiment of the present application also provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any one of the control loop monitoring and diagnosis methods provided above.

[0141] An embodiment of the present application further provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0142] The memory is used to store computer programs;

[0143] The processor is used to execute the computer program so that the electronic device can implement any one of the control loop monitoring and diagnosis methods provided above.

[0144] An embodiment of the present application also provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any one of the control loop monitoring and diagnosis methods provided above.

[0145] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the embodiments of the present application. Therefore, the embodiments of the present application will not be limited to the embodiments shown herein, but will conform to the widest range consistent with the principles and novel features disclosed herein.

Claims

1. A control loop monitoring and diagnosis method, characterized in that: include: Receive a data monitoring request for a specified control loop; wherein there are multiple control loops in the industrial system, each control loop includes multiple nodes, each node is equipped with at least one pin, and the pin of the node is connected to the corresponding pin of other nodes in the same control loop through a wire; a wire relationship list and a node information list are stored in the database of the industrial system for each control loop; the wire relationship recorded in the wire relationship list for each wire in the control loop includes the name of the leading pin of the wire, the ID of the leading node, the name of the trailing pin and the ID of the trailing node; the node information recorded in the node information list for each node in the control loop includes the ID and operation monitoring data of the node; Traversing the connection relationship list corresponding to the specified control loop, searching for any connection relationship containing a target element, wherein the target element is the name of a leading or trailing pin of any connection connected to a predetermined key node in the specified control loop; then, determining the ID of the key node according to the found connection relationship and the inherent correspondence between the key node and its connection; Traversing the node information list corresponding to the specified control loop, searching for the node information containing the ID of the key node, and extracting the operation monitoring data of the node from the found node information; The extracted operation monitoring data is used as input data of a pre-built equipment diagnosis model, and the equipment diagnosis model performs calculations based on the input data to obtain an operation condition diagnosis result of a specified control loop.

2. The control loop monitoring and diagnosis method according to claim 1, characterized in that: The operation monitoring data of the node includes the real-time value and historical value of each pin of the node, the real-time value of the operation parameter of the node and the real-time value of the monitoring parameter of the node.

3. The control loop monitoring and diagnosis method according to claim 1, characterized in that: The operating condition diagnosis results are reflected in three dimensions: control performance of the control loop, instrument health status of the control loop, and oscillation and nonlinear status of the control loop; The control performance of the control loop refers to a comprehensive evaluation of the control effect of the control loop; The instrument health status of the control loop refers to a comprehensive evaluation of the health status of each instrument in the control loop; The oscillation and nonlinear state of the control loop refers to a comprehensive evaluation of abnormal phenomena induced by oscillation or nonlinear factors in the control loop.

4. The control loop monitoring and diagnosis method according to claim 3, characterized in that: The control performance of the control loop is quantified through multiple indicators, including: stable segment control performance indicator, sluggishness indicator, MV stability indicator, PV-SV deviation, environmental interference and MV overlimit percentage; The stable segment control performance index is used to evaluate the fluctuation of the process variable PV around the set value SV; The sluggishness index is used to measure the hysteresis degree of the control loop response; The MV stability index is used to evaluate the stability of the manipulated variable MV in the control loop; The PV-SV deviation refers to the degree of deviation between the process variable PV and the set value SV; The environmental interference degree is used to measure the influence of external environmental factors on the control loop; The MV limit exceeding percentage refers to the proportion of the manipulated variable MV exceeding its allowable range; Each indicator under the control performance of the control loop divides the state of the control loop into three categories: fault, abnormal and normal: when the indicator value exceeds the preset acceptable range, the control loop is judged to be in a fault state; if the indicator value has not reached the fault threshold but has exceeded the normal range, the control loop is judged to be in an abnormal state; and when the indicator value is within the normal range, it indicates that the control loop is in a normal state.

5. The control loop monitoring and diagnosis method according to claim 3, characterized in that: The instrument health status of the control loop is quantified through multiple indicators, including: valve sticking comprehensive index, valve stem slip index, valve stem dead zone index, MV-PV sticking mode index and sensor accuracy index; The valve viscosity comprehensive index is used to evaluate the viscosity of the valve during movement; The valve stem sliding index is used to quantify the frequency and amplitude of valve stem sliding; The valve stem dead zone index is used to quantify the range of the valve stem movement without causing significant changes in the process variable; The MV-PV sticky mode index is used to evaluate the stickiness between the manipulated variable MV and the process variable PV; The sensor accuracy index is used to evaluate the measurement accuracy and stability of the sensor; Each indicator of the instrument health status of the control loop divides the status of the control loop into three categories: fault, abnormal and normal: when the indicator value exceeds the preset acceptable range, the control loop is judged to be in a fault state; if the indicator value has not reached the fault threshold but has exceeded the normal range, the control loop is judged to be in an abnormal state; and when the indicator value is within the normal range, it indicates that the control loop is in a normal state.

6. The control loop monitoring and diagnosis method according to claim 3, characterized in that: The oscillation and nonlinear state of the control loop are quantitatively presented through multiple indicators, including: oscillation indicator, oscillation period indicator, oscillation spectrum power percentage indicator and nonlinear indicator; The oscillation index is used to quantify the oscillation phenomenon that occurs during the operation of the control loop; The oscillation period indicator is used to control the time interval of repeated oscillation phenomena in the loop; The oscillation spectrum power percentage index is used to quantify the relative strength of different frequency oscillation components in the control loop; The nonlinear index is used to quantify the influence of nonlinear characteristics in the control loop; Each indicator under the oscillation and nonlinear state of the control loop divides the state of the control loop into three categories: fault, abnormal and normal: when the indicator value exceeds the preset acceptable range, the control loop is judged to be in a fault state; if the indicator value has not reached the fault threshold but has exceeded the normal range, the control loop is judged to be in an abnormal state; and when the indicator value is within the normal range, it indicates that the control loop is in a normal state.

7. The control loop monitoring and diagnosis method according to claim 4, 5 or 6, characterized in that: Each indicator is presented through a key performance indicator (KPI) historical baseline chart.

8. The control loop monitoring and diagnosis method according to claim 1, characterized in that: The connection relationship set for each connection in the control loop in the connection relationship list also includes: the name of the preceding node and the name of the succeeding node of the node; The node information list includes the node information recorded for each node in the control loop, and further includes: the name of the node; The connection relationship list further includes the connection relationship set for each connection in the control loop: the ID of the front pin and the ID of the rear pin of the node; The database of the industrial system also stores, for each control loop, the ID of the control loop, the type of the control loop, the name of the control loop, and the description of the control loop.

9. The control loop monitoring and diagnosis method according to claim 1, characterized in that: After extracting the operation monitoring data of the node from the found node information, the method further includes: storing the node type as a key node in the found node information; The control loop monitoring and diagnosis method also includes: when a data monitoring request for the same designated control loop is received again, traversing the node information list, searching for node information containing a node type of a key node, and extracting the node's operation monitoring data from the found node information; using the extracted operation monitoring data as input data of a pre-built equipment diagnosis model, the equipment diagnosis model performs calculations based on its input data to obtain an operating condition diagnosis result of the designated control loop.

10. A computer program product, characterized in that It comprises computer-readable instructions, and when the computer-readable instructions are executed on an electronic device, the electronic device implements the control loop monitoring and diagnosis method as claimed in any one of claims 1 to 9.

Images