Method system and system for monitoring, evaluating and optimizing control loop
By monitoring the steady state and adjustment quality of the control loop, the problem of inability to effectively evaluate the control loop regulation quality in the prior art is solved, efficient and rapid adjustment quality improvement is achieved, and production efficiency and safety are improved.
Patent Information
- Application Number
- CN202411233023.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-11-29
AI Technical Summary
The prior art cannot effectively monitor and evaluate the adjustment quality of the control circuit, resulting in insufficient production efficiency and safety.
By monitoring the start time, target value, actual value, control command information and feedback information of the adjusted object, it is determined whether the adjusted object and the control loop are in steady state, and the monitoring information of the adjustment quality is obtained based on these data, and the monitoring information of the adjustment quality is carried out. Quantitative evaluation and optimization tips.
It realizes automated control loop regulation monitoring and quality evaluation, improves production efficiency and safety, and can quickly improve the adjustment quality.
Smart Images

Figure CN119045462A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of electronic technology, and in particular to a control loop monitoring, evaluation, and optimization method system and system. Background Art
[0002] A large number of control loops (including open-loop control loops and closed-loop control loops, etc.) are used in modern factories or production facilities. For example, in a power plant, its generator set can be controlled by a control loop to ensure normal power generation. These control loops can be applied to, but not limited to: gas turbine control, steam turbine control, waste heat boiler control and other auxiliary equipment control.
[0003] With the widespread application of control loops, the control and regulation effect of the controlled objects directly affects production efficiency and production safety. Therefore, how to efficiently and accurately monitor and evaluate the control quality of control loops has become an important issue to ensure production efficiency and safety. Summary of the invention
[0004] In order to solve the above technical problems, an embodiment of the present application provides a control loop monitoring, evaluation, and optimization method system and system to solve at least one of the above problems.
[0005] According to a first aspect of an embodiment of the present application, a method for monitoring a control loop is provided, the method being applied to a device provided with a control loop and a regulated object, the control loop comprising a controller and an actuator, the controller controlling the action of the actuator to regulate the regulated object, the method comprising: determining a start time for the control loop to regulate the regulated object, and obtaining a target value and an actual value of a regulated parameter of the regulated object and a time when the actual value appears, as well as control instruction information and feedback information of the actuator; determining whether the regulated object and the control loop are in a steady state according to the actual value of the regulated parameter of the regulated object and the time when the actual value appears, the control instruction information and the feedback information; if in a steady state, determining monitoring information on the regulation quality of the control loop according to at least part of the start time of regulation, the actual value and the time when the actual value appears, the target value, the control instruction information and the feedback information.
[0006] According to a second aspect of an embodiment of the present application, a control loop evaluation method is provided, comprising: obtaining monitoring information for indicating the quality of regulation of a regulated object by the control loop, the monitoring information being obtained by the above method, the monitoring information including at least one of the following parameters: regulation duration, oscillation duration, regulation deviation, overshoot value and variance; obtaining benchmark information and weights corresponding to the monitoring information, the benchmark information including at least one of the following benchmark values: regulation duration benchmark value, oscillation duration benchmark value, regulation deviation benchmark value, overshoot benchmark value and variance benchmark value; determining a monitoring score of the control loop based on the monitoring information, preset benchmark information and weights.
[0007] According to a third aspect of an embodiment of the present application, a control loop optimization method is provided, comprising: obtaining a monitoring score and a deviation score of the control loop, the monitoring score and the deviation score being obtained by the above method; and determining optimization prompt information based on the monitoring score, the deviation score and preset optimization conditions.
[0008] Through the monitoring, evaluation, and optimization method system of the control loop of the embodiment of the present application, by obtaining the start time of the control loop's regulation of the regulated object, as well as the target value, control instruction information, and feedback information, and determining whether the regulated object and the control loop are in a steady state based on the acquired data, when both are in a steady state, the monitoring information used to characterize the regulation quality of the control loop is determined, thereby realizing automated monitoring and quality evaluation of the regulation of the control loop, solving the problem that the regulation quality of the control loop cannot be determined in the prior art, resulting in insufficient production efficiency and safety. In addition, through effective monitoring of the control loop's regulation quality, it can be quantitatively evaluated based on the monitoring information, and then the regulation process with poor regulation quality can be analyzed based on the quantitative evaluation results and the monitoring information, and then the optimization prompt information is output, thereby realizing efficient and rapid regulation quality improvement. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Figure 1 It is a schematic diagram of the connection of a device, a control loop and a monitoring system provided in an embodiment of the present application;
[0010] Figure 2 It is a schematic diagram of the steps of the control loop monitoring method provided in the embodiment of the present application;
[0011] Figure 3 It is a schematic diagram of the steps of the control loop evaluation method provided in the embodiment of the present application;
[0012] Figure 4 It is a schematic diagram of the steps of the control loop optimization method provided in the embodiment of the present application;
[0013] List of reference numerals:
[0014] 10. Equipment; 11. Controlled object; 12. Control loop; 121. Controller; 122. Actuator; 20. Monitoring system of the control loop.
[0015] Step S201: determine the start time of the control loop to adjust the regulated object, and obtain the target value and actual value of the regulated parameter of the regulated object and the time when the actual value appears, as well as the control instruction information and feedback information of the actuator.
[0016] Step S202: determining whether the regulated object and the control loop are in a steady state according to the actual value of the regulated parameter of the regulated object and the time when the actual value appears, the control instruction information and the feedback information.
[0017] Step S203: If in a steady state, the monitoring information of the regulation quality of the control loop is determined according to at least part of the regulation start time, the actual value and the time when the actual value appears, the target value, the control instruction information and the feedback information.
[0018] Step S301: Acquire monitoring information indicating the quality of regulation of a regulated object by a control loop.
[0019] Step S302: Obtain benchmark information and weights corresponding to the monitoring information.
[0020] Step S303: Determine the monitoring score of the control loop according to the monitoring information, preset benchmark information and weight.
[0021] Step S401: Obtain the monitoring score and deviation score of the control loop.
[0022] Step S402: Determine optimization prompt information according to the monitoring score, the deviation score and preset optimization conditions. DETAILED DESCRIPTION
[0023] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.
[0024] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.
[0025] Reference Figure 1 For ease of understanding, before explaining the monitoring method of the control loop and the working process of the system, a brief example explanation of the application scenario and working process of the control loop is given.
[0026] The method can be applied to any device 10 equipped with a control loop 12, wherein the device 10 includes a controlled object 11, and the control loop 12 includes a controller 121 and an actuator 122, wherein the controller 121 controls the actuator 122 to adjust the controlled object 11. The controlled object 11, the controller 121, and the actuator 122 may be different for different devices 10 used. The controller 121 may receive manually input instructions and output control instructions to the actuator 122, or the controller 121 may be a PID regulator, which sends control instructions to the actuator 122 based on the monitored controlled parameters of the controlled object 11 or feedback information fed back by the actuator 122.
[0027] The control loop 12 may be any appropriate open-loop control loop or closed-loop control loop. It may be a single-variable control loop (one adjusted parameter) or a complex control loop (including multiple actuators). For a complex control loop, it may be split into multiple single-variable control loops for monitoring, or the differences between the actuators 122 may be ignored and the time axes of multiple actuators 122 may be merged for monitoring.
[0028] In the application scenario of a power plant, the equipment may be a gas turbine, a steam turbine, etc. The control loop may be a gas turbine control system loop, a steam turbine control system loop, a waste heat boiler control system loop, and an auxiliary control system loop.
[0029] For gas turbines, the following controls may be performed, but are not limited to: start / speed increase control, power / load / speed control (including rate limit control), turbine exhaust temperature control, combustion chamber pressure pulsation control, compressor compression ratio control, IGV (inlet guide vane) adjustable inlet guide vane control, fuel temperature and pressure control (including: control of fuel valves such as premixing valve and duty valve, and control of FGH fuel gas heater fuel preheating system, gas pressure regulating valve, etc.), turbine body temperature control (TCA turbine cooling air system), etc.
[0030] For the steam turbine, the following controls may be performed on it, but are not limited to: steam turbine speed control, condenser vacuum control, etc.
[0031] For the waste heat boiler part, the following controls can be performed on it, but not limited to: water level control, pressure control, wall temperature control of the air bag, and related control loops of the high-pressure steam control system, medium-pressure steam control system, low-pressure steam control system, etc. on the waste heat boiler side.
[0032] Other auxiliary control system loops can perform the following controls, but are not limited to: fixed cooling pump control, open cooling pump control, closed cooling pump control, main lubricating oil system control, SCR and other denitrification system control, sewage pump control, rainwater pump control, etc.
[0033] For ease of understanding, some nouns are explained as follows:
[0034] Actual value: recorded as PV, which is the value of the monitored parameter.
[0035] Target value: denoted as SP, which is the value that the adjusted parameter needs to achieve.
[0036] First error: It can be set as needed, which represents the allowable error value when the adjusted parameter is in a steady state, such as ±1%.
[0037] The first steady-state interval: It can be determined according to the first error and the target value. For example, if the first error is ±1% and the target value is 3000 rpm, the first steady-state interval is 2970 rpm to 3030 rpm.
[0038] First duration: It can be set as needed, such as 3 minutes.
[0039] Control instruction information: denoted as CO. The control instruction information includes the target state value and time. The target state value is usually in the form of percentage, such as 30%, 50%, etc.
[0040] Feedback information: denoted as FB, feedback information includes actual state value and time. The actual state value is usually in the form of percentage, such as 45%.
[0041] Second error: It can be set as needed, indicating the allowable error value when the actuator is in a steady state, such as ±5%.
[0042] The second steady-state interval: It can be determined according to the second error and the target state value. For example, if the second error is ±3% and the target state value is 20%, the second steady-state interval is 19.4%~20.6%.
[0043] Second duration: It can be set as needed, such as 3 seconds.
[0044] Combine the following Figure 2, a control loop monitoring method provided in an embodiment of the present application is described. In the process of the control loop adjusting the adjusted parameters of the adjusted object, the method can monitor the adjustment process and evaluate its adjustment quality, so as to facilitate the staff to grasp the working quality of the control loop in time, thereby ensuring the safety and production efficiency of the equipment production process. The method includes the following steps:
[0045] Step S201: determine the start time of the control loop to adjust the regulated object, and obtain the target value and actual value of the regulated parameter of the regulated object and the time when the actual value appears, as well as the control instruction information and feedback information of the actuator.
[0046] By monitoring the data in the control loop, the start time of the adjustment process, the target value of the adjusted parameter, one or more actual values (each actual value corresponds to the time of its occurrence), control instruction information and feedback information can be obtained. For example, the execution device of the monitoring method (such as a computer equipped with the method) can obtain the aforementioned data from the control loop, or it can obtain the aforementioned data by connecting a separately configured sensor, reading the data collected by the sensor, and analyzing the data, without limitation.
[0047] In some examples, determining the start time of the control loop to adjust the regulated object includes: determining the time when the target value of the regulated parameter configured in the control loop is changed to a new target value as the start time of the adjustment. For example, when the user changes the target value from 3000 rpm to 5000 rpm, this time can be determined as the start time of the adjustment process.
[0048] Or, in other examples, the moment when the actual value of the adjusted parameter of the adjusted object exceeds the set interval is determined as the adjustment start time. If the device has been operating normally for a period of time, in this case, the actual value of the adjusted parameter is maintained within the first steady-state interval. In this case, if the actual value of the adjusted parameter exceeds the first steady-state interval (the first steady-state area can be considered as the set interval), the controller starts a new adjustment process, and this moment is confirmed as the adjustment start time.
[0049] Step S202: determining whether the regulated object and the control loop are in a steady state according to the actual value of the regulated parameter of the regulated object and the time when the actual value appears, the control instruction information and the feedback information.
[0050] Wherein, if the duration of the actual value being in the first steady-state interval satisfies the first duration, it is determined that the regulated object is in a steady state, wherein the first steady-state interval is determined according to the target value and the set first error. For example, the first steady-state interval is 2970 rpm to 3030 rpm. During a certain regulation process, if the actual value is maintained in the first steady-state interval for 3 consecutive minutes for the first time, it can be determined that the regulated object is in a steady state, and this moment is recorded as the second moment. For example, the actual value monitored at moment A is 2971, and the first duration is 3 minutes, then the actual values monitored in the following 3 minutes are all between 2970 rpm and 3030 rpm, and it is determined that the regulated object is in a steady state.
[0051] or,
[0052] If the duration of the actual state value of the actuator indicated by the feedback information being in the second steady-state interval satisfies the second duration, it is determined that the control loop is in a steady state, wherein the second steady-state interval is determined based on the target state value and a preset second error. For example, the second steady-state interval is 19.4% to 20.6%. During a certain adjustment process, if the actual state value is maintained in the second steady-state interval for 3 consecutive seconds for the first time, it can be determined that the control loop is in a steady state, and this moment is recorded as the third moment. For example, at moment B, it is monitored that the actual state value of the feedback information is in the second steady-state interval, and the second duration is 3 seconds. In the following 3 seconds, the actual state value is in the second steady-state interval, and it is determined that the control loop is in a steady state.
[0053] The steady-state moment of the regulated object and the control loop as a whole can be the later of the second moment and the third moment.
[0054] Of course, different steady-state determination conditions may be set according to different needs, and there is no limitation to this.
[0055] If the regulated object still fails to reach a steady state after exceeding a first timeout threshold (such as 5 minutes), the method may terminate the current monitoring and evaluation and output a prompt that the regulated object cannot reach a steady state or other appropriate prompt information.
[0056] Alternatively, if the regulated object reaches a steady state, but exceeds a second timeout threshold (such as 3 minutes or 5 minutes, etc.), and the control loop still cannot reach a steady state, the method can terminate this monitoring and evaluation, and output a prompt that the control loop cannot reach a steady state or other appropriate prompt information.
[0057] Alternatively, if both have reached a steady state, the following step S203 may be executed.
[0058] Step S203: If it is in a steady state, the monitoring information of the regulation quality of the control loop is determined according to at least part of the regulation start time, the actual value and the corresponding time, the target value, the control instruction information and the feedback information.
[0059] Depending on the needs or configurations, the monitoring information may include different contents. In the present embodiment, it may include one of the following but is not limited to: adjustment duration, oscillation duration of the control loop, adjustment deviation, overshoot value, variance of the adjusted parameter, control time constant, damping time of the adjusted object, damping time of the actuator, utilization rate, etc.
[0060] Regarding the adjustment duration: based on the actual value and the corresponding time, the target value, the control instruction information and the feedback information, determine the steady-state moment when the control loop and the regulated object are in a steady state, and based on the steady-state moment and the adjustment start time, determine the adjustment duration of the control loop.
[0061] For example, adjustment duration = steady-state moment - adjustment start moment. As mentioned above, the steady-state moment can be the later of the second moment and the third moment. The second moment can be the moment when the regulated object reaches a steady state, and the third moment can be the moment when the control loop reaches a steady state.
[0062] Regarding the oscillation duration of the control loop: determine the first adjustment oscillation duration of the regulated object based on the actual value and the corresponding time as well as the target value, and determine the second adjustment oscillation duration of the actuator based on the control instruction information and the feedback information; and determine the oscillation duration of the control loop based on the first adjustment oscillation duration and the second adjustment oscillation duration.
[0063] Among them, the first regulation oscillation duration can be specifically determined in the following manner: determine the first moment when the actual value reaches the first steady-state interval, and the first steady-state interval is determined according to the target value and the set first error; determine the second moment when the duration of the actual value in the first steady-state interval satisfies the first duration; and use the duration between the first moment and the second moment as the first regulation oscillation duration.
[0064] For example, the moment when the actual value first reaches the first steady-state interval is the first moment. As in the previous example, during the adjustment process, the actual value will still change after first reaching the first steady-state interval. During the change process, it may remain in the first steady-state interval or may exceed the first steady-state interval. This method continuously monitors the actual value until the duration of the actual value remaining in the first steady-state interval meets the first duration. This moment is recorded as the second moment. At this time, it can be determined that the regulated object is in a steady state, that is, the oscillation of the regulated object ends.
[0065] The first adjustment oscillation duration (which may be referred to as the working condition oscillation duration) = the second moment - the first moment.
[0066] The second regulation oscillation duration can be specifically determined in the following manner: determine the third moment when the duration of the actual state value of the actuator indicated by the feedback information in the second steady-state interval satisfies the second duration, and the second steady-state interval is determined according to the target state value indicated by the control instruction information and the preset second error; if the third moment is earlier than or equal to the second moment, the second regulation oscillation duration is 0. When the third moment is earlier than or equal to the second moment, it means that the actuator enters the steady state earlier than the regulated object, and the second regulation oscillation duration is 0; if the third moment is later than the second moment, the second regulation oscillation duration is the duration between the third moment and the second moment.
[0067] For example, if the actual state value is maintained within the second steady-state interval for a duration that satisfies the second duration (such as 3 consecutive seconds), it indicates that the actuator enters a steady state (ie, the control loop enters a steady state), and this moment is recorded as the third moment.
[0068] If the third moment is earlier than or equal to the second moment, the second adjustment oscillation duration is 0. If the third moment is later than the second moment, it means that after the oscillation of the regulated object ends, the actuator is still oscillating. In this case, the second adjustment oscillation duration = the third moment - the second moment.
[0069] The oscillation duration of the control loop = the first adjustment oscillation duration + the second adjustment oscillation duration.
[0070] Regarding the regulation deviation: the regulation deviation is determined according to the difference between the actual value and the target value when the regulated object and the control loop are in a steady state.
[0071] If the controlled object reaches steady state later than the control loop (i.e. the second moment is later than the third moment), the absolute value of the difference between the actual value of the controlled parameter when the controlled object reaches steady state and the target value is used as the control deviation.
[0072] Alternatively, if the regulated object reaches steady state earlier than the control loop (i.e., the second moment is earlier than the third moment), the absolute value of the difference between the actual value of the regulated parameter and the target value when the control loop reaches steady state is used as the regulation deviation.
[0073] Regarding the overshoot value: when the actual value and the target value overlap, the overshoot value of the control loop is determined based on the actual value and the target value.
[0074] For example, in the process of the regulated object reaching a steady state, the coincidence of the actual value and the target value can be understood as the actual value being equal to the target value. In this case, if there is a situation where the actual value exceeds the first steady-state interval, the maximum value of the actual value exceeding the target value is taken as the overshoot value. In a specific example, the actual value of the regulated object is equal to the target value at time t5, and it is not until time t6 that the regulated object is determined to have reached a steady state. During the time period from time t5 to time t6, N actual values may be sampled. If there are M actual values among the N actual values that exceed the first steady-state interval, the actual value with the largest difference from the target value among the M actual values is taken, and the difference between the actual value with the largest difference and the target value is calculated as the overshoot value, and both N and M are greater than or equal to 0.
[0075] Regarding the variance of the regulated parameter: the variance of the regulated parameter in the oscillation phase of the regulated object and the control loop is determined according to the actual value and the corresponding time and the target value.
[0076] If the controlled object reaches steady state earlier than the control loop (i.e. the second moment is earlier than the third moment), then P actual values are sampled between the first moment and the third moment, the differences between the P actual values and the target values are calculated, and the variance is calculated based on the P differences. The variance is used as the variance of the controlled parameter, and P is greater than or equal to 0.
[0077] Alternatively, if the controlled object reaches steady state later than the control loop (i.e., the second moment is later than the third moment), then Q actual values are sampled between the first moment and the second moment, the differences between the Q actual values and the target values are calculated, and the variance is calculated based on the Q differences, and the variance is used as the variance of the controlled parameter, where Q is greater than or equal to 0.
[0078] Regarding the control time constant: the control time constant is the time between the start time of the adjustment and the fourth time, and the fourth time is the time when the actuator starts to act as indicated by the feedback information. The control time constant includes the controller operation time (i.e., PID operation time), signal conversion and transmission time, actuator operation and action time, and the control time constant is included in the adjustment duration.
[0079] Regarding the damping time of the regulated object: the damping time of the regulated object is the time between the first moment and the second moment. The first moment is the moment when the actual value reaches the first steady-state interval (the aforementioned reaching can be understood as the first reaching), and the second moment is the moment when the duration of the actual value in the first steady-state interval satisfies the first duration.
[0080] The damping time is used to characterize the time required for the actual value of the regulated parameter of the regulated object to reach the first steady-state interval for the first time and to be adjusted to the steady state. To a certain extent, it indicates the adjustment speed of the regulated object.
[0081] Regarding the damping time of the actuator: the damping time of the actuator is the time between the fifth moment and the third moment, the fifth moment is the moment when the actual state value indicated by the feedback information reaches the second steady-state interval (the aforementioned reaching can be understood as the first reaching), and the third moment is the moment when the duration of the actual state value in the second steady-state interval satisfies the second duration.
[0082] By determining the control time constant, the damping time of the controlled object, and the damping time of the actuator, the performance and quality of the controller, actuator, etc. of the control loop are effectively monitored.
[0083] The damping time of the actuator is used to characterize the time required for the actual state value of the actuator to reach the second steady-state interval for the first time and to adjust to the steady state. To a certain extent, it characterizes the adjustment speed of the actuator.
[0084] Regarding the utilization rate: based on the switching signal of the switch corresponding to the control loop and at least one of the actual value of the regulated object, determine the first working time of the control loop in the working state of regulating the regulated object and the second working time of the equipment in the working state; based on the first working time and the second working time, determine the utilization rate of the control loop.
[0085] The utilization rate is used to characterize the ratio of the time the control loop participates in the control (i.e., the first working time) to the working time of the equipment (the second working time) during the operation of the equipment.
[0086] In this embodiment, the first working time can be determined by directly detecting the switch signal of the switch corresponding to the control loop, and the second working time can be determined by detecting the switch signal of the switch of the device. The first working time and the second working time can also be determined indirectly by detecting some other parameters, for example, the first working time can be determined by detecting the actual value of the regulated object corresponding to the control loop, and there is no limitation on this. The detected signal can be a digital signal or an analog signal, and there is no limitation on this.
[0087] The control loop can be reliably and accurately monitored in the above manner, and relevant monitoring information of the regulation quality can be obtained. The parameters in the above monitoring information can be displayed to the user as output information according to the user's needs, so as to facilitate the user to view.
[0088] Reference Figure 3 , shows a schematic flow chart of the steps of a control loop evaluation method according to an embodiment of the present application. The evaluation method comprises the following steps:
[0089] Step S301: Acquire monitoring information indicating the quality of regulation of a regulated object by a control loop.
[0090] In some examples, the monitoring information can be obtained by the aforementioned monitoring method, and the monitoring information includes at least one of the following parameters: adjustment duration, oscillation duration, adjustment deviation, overshoot value, and variance. According to different evaluation requirements, the monitoring information may also include other parameters, which are not limited.
[0091] Step S302: Obtain benchmark information and weights corresponding to the monitoring information.
[0092] The benchmark information includes at least one of the following benchmark values: an adjustment duration benchmark value, an oscillation duration benchmark value, an adjustment deviation benchmark value, an overshoot benchmark value, and a variance benchmark value. It should be noted that the parameters in the monitoring information should correspond to the benchmark values in the benchmark information. If the monitoring information includes the adjustment duration, the benchmark information should include the adjustment duration benchmark value.
[0093] The aforementioned adjustment duration reference value, oscillation duration reference value, adjustment deviation reference value, overshoot reference value and variance reference value can be obtained by machine learning based on historical adjustment data. For example, for the adjusted objects of the same type or the same working scene, the adjustment data of a period of history (including but not limited to the historical actual value, target value, control instruction information and feedback information, etc.) are collected. The user can select the data with better or more satisfactory adjustment effect from the historical adjustment data as training samples as needed, input the training samples into the machine learning model, and obtain the adjustment duration reference value, oscillation duration reference value, adjustment deviation reference value, overshoot reference value and variance reference value learned by the machine learning model.
[0094] Of course, the method of obtaining these reference values is not limited to the aforementioned methods, and other methods may also be used, such as manual selection, etc., which is not limited to this.
[0095] The weights include the weights corresponding to the various parameters of the monitoring information. For a clear description, the weight corresponding to the adjustment duration is recorded as weight B1, the weight of the oscillation duration is recorded as weight B2, the weight of the adjustment deviation is recorded as weight B3, the weight of the overshoot value is recorded as weight B4, and the weight of the variance is recorded as weight B5.
[0096] Step S303: Determine the monitoring score of the control loop according to the monitoring information, preset benchmark information and weight.
[0097] In some optional examples, step S303 can be implemented as follows: for each parameter in the monitoring information, determine the deviation score of the parameter based on the difference ratio between the parameter and the corresponding baseline value, calculate the product of the deviation score of the parameter and the corresponding weight as the weighted result of the parameter; determine the monitoring score based on the sum of the weighted results of each parameter.
[0098] Since the processes of determining the deviation scores and weighted results of each parameter are similar, the present embodiment takes the adjustment duration as an example to illustrate the determination process, and the determination processes of other parameters are not repeated.
[0099] When calculating the difference ratio of the adjustment duration, first calculate the difference between the adjustment duration in the monitoring information and the adjustment duration benchmark value, and then divide the difference by the adjustment duration benchmark value to obtain the difference ratio of the adjustment duration. Determine the corresponding deviation score based on the difference ratio. If the difference ratio of the section duration is less than or equal to 3%, the deviation score is 100 points; if it is greater than 3% and less than or equal to 5%, the deviation score is 95 points; if it is greater than 5% and less than or equal to 10%, the deviation score is 90 points, and so on. The corresponding relationship between the difference ratio and the deviation score can be set and adjusted as needed. Multiply the deviation score of the adjustment duration and the weight B1 of the adjustment duration, and the product result is used as the weighted result of the adjustment duration.
[0100] After the weighted results of each parameter are calculated, the weighted results of each parameter are added together to obtain the monitoring score.
[0101] It should be noted that the parameters involved in the monitoring score can be determined by a user-defined configuration or based on a default configuration.
[0102] In this way, the regulation quality of the control loop can be scored based on one or more parameters in the monitoring information, allowing the user to intuitively determine the regulation quality. In this way, the regulation quality can be effectively quantified, and the results can be output to the user, allowing him to intuitively determine the regulation quality.
[0103] Reference Figure 4 , shows a schematic flow chart of a control loop optimization method. The method comprises the following steps:
[0104] Step S401: Obtain the monitoring score and deviation score of the control loop.
[0105] The monitoring score and deviation score can be obtained through the aforementioned evaluation method.
[0106] Step S402: Determine optimization prompt information according to the monitoring score, the deviation score and preset optimization conditions.
[0107] In some examples, for situations where the monitoring score is low, such as below a set score threshold (eg, 80 points or other score thresholds set as required), a prompt message may be output to facilitate maintenance or optimization by staff.
[0108] For example, the optimization condition includes at least one of the following: an adjustment timeout condition, a deviation condition, and an oscillation condition. Accordingly, step S402 may include at least one of the following:
[0109] Case A: If it is determined that the adjustment timeout condition is met based on the monitoring score and the deviation score, a first optimization prompt information is output, and the first optimization prompt information indicates to perform the following operations in sequence: shorten the integral time, reduce the proportional band, and increase the differential time.
[0110] In some cases, if the monitoring score is lower than the score threshold, and the deviation score of the adjustment time is lower than the set threshold (such as 90 points), it can be determined that the adjustment time meets the adjustment timeout condition (other adjustment timeout conditions can also be set as needed, which is not limited in this embodiment), indicating that there is a problem of slow adjustment. In this case, for the manual adjustment method, a prompt message can be output to the user. The prompt message indicates that the user can be advised to optimize by appropriately shortening the integral time (i.e., adjusting I in PID control), appropriately reducing the proportional band (i.e., adjusting P in PID control), and appropriately increasing the differential time (i.e., adjusting D in PID control). The aforementioned optimization steps have priorities and the order is as above. The prompt message can indicate the recommended optimization operation of the next step, which will only be performed if the previous step is not effective or the adjustment purpose is not achieved. In special or emergency situations, the operation order of each of the above situations can be flexibly adjusted.
[0111] For the automatic tuning method, this method can screen out executable quantitative standards (i.e., the aforementioned parameters) based on the historical operating data of the control loop, and compare them with the real-time data to obtain the real-time residual between the two. When the residual exceeds the set interval, the automatic tuning of the PID parameters is triggered, and the P, I, and D parameters in the PID control are adjusted in the order of steps indicated by the aforementioned prompt information.
[0112] Case B: If it is determined that the deviation condition is met based on the monitoring score and the deviation score, a second optimization prompt information is output, and the second optimization prompt information indicates to perform the following operations in sequence: reduce the proportional band, increase the integral time, and increase the differential time.
[0113] In some cases, if the monitoring score is lower than the score threshold, and the deviation score of the adjustment deviation is lower than the set threshold (such as 90 points), it can be determined that the adjustment deviation meets the deviation condition (the set threshold of the deviation condition can also be modified as needed, without restriction), indicating that there is a problem of excessive deviation. For manual tuning, prompt information can be output to indicate the following operations in sequence: appropriately reduce the proportional band (i.e. adjust P in PID control), appropriately increase the integral time (i.e. adjust I in PID control), and appropriately increase the differential time (i.e. adjust D in PID control). The priority prompt description is similar to the previous one, so it will not be repeated.
[0114] For the automatic adjustment mode, the processing process is similar to the previous case where the adjustment time is too long, so it will not be repeated here.
[0115] Case C: If it is determined that the oscillation condition is met according to the monitoring score and the deviation score, a third optimization prompt information is output, and the third optimization prompt information indicates to perform the following operations in sequence: increase the integral time, increase the differential time, and enlarge the proportional band.
[0116] In some cases, if the monitoring score is lower than the score threshold, and the deviation score of the oscillation duration is lower than the set threshold (such as 90 points), it can be determined that the oscillation duration meets the oscillation condition, indicating that the oscillation is serious. If manual tuning is used, a prompt message can be output to indicate the following operations in sequence: appropriately increase the integral time (i.e. adjust I in PID control), appropriately increase the differential time (i.e. adjust D in PID control), and appropriately enlarge the proportional band (i.e. adjust P in PID control). The priority prompt description is similar to the previous one, so it will not be repeated.
[0117] For the automatic tuning mode, the processing process is similar to the previous one, so it will not be repeated here.
[0118] The method can monitor the regulation quality of the control loop and timely discover problems such as insufficient control regulation accuracy, slow regulation rate, regulation oscillation, etc. in the control loop during the production process of existing industrial equipment, so that users can quickly solve the above problems and avoid serious accidents such as reduced work efficiency and equipment service life, abnormal equipment shutdown, and casualties caused by the existence of these problems.
[0119] The above-mentioned monitoring, evaluation and optimization method system can monitor the data of the control loop in a timely and accurate manner, analyze and process the data, so as to identify the problems of slow or over-fast adjustment, adjustment oscillation (generally caused by overshoot), and low adjustment accuracy (large error) in the adjustment process of the control loop. In addition, these methods can also monitor and evaluate the quality of the control loop itself, such as the algorithm level of the controller, the control / feedback unit of the field actuator (for example: the servo, positioner, inverter, etc. of the valve), the actuator unit (for example: the motor, gearbox, torque of the opening / closing stroke, baffle of the electric valve, the cylinder, piston, spring, valve stem, valve core, etc. of the pneumatic valve), the hardware of DCS / TCS, etc. (IO baseboard, link and card / channel) and the field working conditions (mechanical characteristics of the controlled variable and the adjustment variable, the status of specific equipment such as pipelines / heat exchangers).
[0120] By configuring the optimal performance index (i.e. the aforementioned benchmark value) for the control loop, quantifying and scoring each adjustment result, and outputting the index evaluation results and cause analysis, users can optimize the control loop.
[0121] According to another aspect of the present application, a monitoring system 20 for a control loop is provided. The monitoring system 20 is connected to the device 10 equipped with the control loop 12 , and the monitoring system 20 is used to execute the above method.
[0122] In some examples, the monitoring system may be a device configured with a data computing unit (such as a CPU or a cloud device, etc.), and the monitoring system may share a data computing unit with the device where the control loop is located, or may be a different data computing unit, without limitation.
[0123] It should be noted that although the above-mentioned system is named as a monitoring system, if necessary, the system can not only realize the function of acquiring monitoring information, but also realize the functions corresponding to the aforementioned evaluation and optimization methods.
[0124] The following is an explanation of the implementation process of the aforementioned control loop monitoring, evaluation, and optimization methods in conjunction with a specific work scenario to facilitate understanding.
[0125] In this working scenario, the device is a gas turbine, the regulated object is the main wheel of the gas turbine, and the regulated parameter is the speed of the main wheel. The control loop can regulate and control the speed of the main wheel. The monitoring system 20 configured with this method can be connected to the control loop, obtain data from the control loop, and process the data to monitor the regulation quality of the control loop.
[0126] In the example of this working scenario, the monitoring system 20 determines several parameters of the control loop based on the acquired data, such as the adjustment time, oscillation time, adjustment deviation, overshoot value, variance of the adjusted parameter, etc., and performs an evaluation based on these parameters to obtain a monitoring score, and then gives optimization prompt information based on the monitoring score.
[0127] It should be noted that this working scenario is explained by taking the monitoring system 20 obtaining data from the control loop to achieve monitoring as an example. In other working scenarios, the monitoring system 20 can be configured with sensors, and the sensors can be installed in suitable locations and obtain data through the sensors, and the aforementioned required parameters can be obtained by analyzing the data. There is no limitation to this.
[0128] The following is an exemplary description of the process of the monitoring system 20 acquiring various parameters:
[0129] Regarding the adjustment duration, it indicates the time required for the control loop to adjust the regulated object once. Adjustment duration = steady-state time - adjustment start time. Among them, the steady-state time is the time when the regulated object and the control loop reach a steady state during this adjustment, and the adjustment start time is the time when the control loop starts a certain adjustment.
[0130] Regarding the start time of adjustment, for example, when the gas turbine switches from the shutdown state to the startup state, the time when the target value of the rotation speed of the main wheel is adjusted from 0 to 3000 rpm can be determined as the start time of adjustment. For another example, when the main wheel is in normal steady-state operation, the time when the staff adjusts the target value from 3000 rpm to 5000 rpm can also be determined as the start time of adjustment. For another example, when the main wheel is in normal steady-state operation, but the rotation speed of the main wheel suddenly exceeds the steady-state range can also be determined as the start time of adjustment.
[0131] Regarding the steady-state moment, for ease of understanding, the adjustment process of the control loop is briefly described: when the control loop adjusts the regulated object, the controller of the control loop sends control instruction information to the actuator connected to the regulated object (including the actuator's target state value and time, for example, the target state value is 30%), and the actuator acts according to the control instruction information to change the actual value of the regulated parameter until both the regulated object and the control loop reach a steady state. In addition, the actuator sends feedback information to the controller (including the actuator's actual state value and time, for example, the actual state value is 28%).
[0132] In this working scenario, whether the regulated object has reached a steady state can be determined in the following manner: the monitoring system 20 determines the first steady-state interval, such as 2970 rpm to 3030 rpm, based on the target value and the preset first error. By comparing the actual value with the first steady-state interval, it can be determined whether the actual value is within the first steady-state interval. The monitoring system 20 records the moment when the actual value first reaches the first steady-state interval as the first moment. If at moment A, the monitoring system 20 determines that the actual value has been in the first steady-state interval for 3 consecutive minutes, it is determined that the regulated object is in a steady state, and moment A is recorded as the second moment.
[0133] Similarly, the method for determining whether the control loop has reached a steady state is, for example, that the monitoring system 20 determines the second steady state interval, such as 19.4% to 20.6%, based on the target state value and the preset second error. Through the actual state value in the feedback information and the time of its occurrence, it can be determined that the duration of the actual state value in the second steady state interval meets the third moment when the second duration is satisfied (for example, if it is determined at moment B that the actual state value has been maintained in the second steady state interval for 3 consecutive seconds, then moment B is recorded as the third moment), and this moment determines that the control loop has reached a steady state.
[0134] When it is determined that the control loop and the regulated object are both in a steady state, the later of the second and third moments is used as the steady state moment. Steady state moment - regulation start moment = regulation duration.
[0135] As for the oscillation duration of the control loop, it is the sum of the first adjustment oscillation duration and the second adjustment oscillation duration.
[0136] The first regulation oscillation duration can be called the operating condition oscillation duration, that is, the time from the actual value of the regulated object reaching the first steady-state interval for the first time to the regulated object reaching the steady state, which can be expressed as: first regulation oscillation duration = second moment - first moment.
[0137] If the control loop reaches steady state later than the regulated object, that is, the third moment is later than the second moment, the second regulation oscillation duration = the third moment - the second moment. If the third moment is earlier than or equal to the second moment, the second regulation oscillation duration is 0.
[0138] Regarding the adjustment deviation, it indicates the deviation value of this adjustment.
[0139] If the control loop reaches steady state later than the regulated object (i.e. the third moment is later than the second moment), the absolute value of the difference between the actual value of the regulated parameter when the control loop reaches steady state and the target value is subtracted, and the regulation deviation is taken as the absolute value of the difference.
[0140] If the control loop reaches steady state earlier than the regulated object (i.e. the third moment is earlier than the second moment), the absolute value of the difference between the actual value of the regulated parameter when the regulated object reaches steady state and the target value is used as the regulation deviation.
[0141] Regarding the overshoot value, it indicates the maximum oscillation degree of this adjustment process.
[0142] For example, the monitoring system 20 analyzes the data to determine that the actual value is equal to the target value at time t5, until the aforementioned second time moment determines that the regulated object has reached a steady state. During the time period from time t5 to the second time moment, N actual values may be sampled. If there are M actual values among the N actual values that exceed the first steady-state interval, then the actual value with the largest difference from the target value among the M actual values is taken, and the difference between the actual value with the largest difference and the target value is calculated as the overshoot value, and both N and M are greater than or equal to 0.
[0143] About the variance of the adjusted parameter:
[0144] If the controlled object reaches steady state earlier than the control loop (i.e. the second moment is earlier than the third moment), then P actual values are sampled between the first moment and the third moment, the differences between the P actual values and the target values are calculated, and the variance is calculated based on the P differences. The variance is used as the variance of the controlled parameter, and P is greater than or equal to 0.
[0145] Alternatively, if the controlled object reaches steady state later than the control loop (i.e., the second moment is later than the third moment), then Q actual values are sampled between the first moment and the second moment, the differences between the Q actual values and the target values are calculated, and the variance is calculated based on the Q differences, and the variance is used as the variance of the controlled parameter, where Q is greater than or equal to 0.
[0146] Based on the analyzed parameters such as adjustment time, oscillation time, adjustment deviation, overshoot value, variance of the adjusted parameter, etc., monitoring information representing the quality of this adjustment can be generated to facilitate the staff to view and understand the adjustment quality of the control loop.
[0147] In addition, the monitoring system 20 can score accordingly, such as comparing each parameter with the corresponding benchmark value and calculating the corresponding score, and then weighted summing the score of each parameter to obtain the final monitoring score, which can intuitively reflect the regulation quality.
[0148] The monitoring system 20 can give optimization prompt information based on the monitoring score. For example, when the monitoring score is less than 80 points, it means that the quality of the control loop adjustment is poor and the control loop needs to be adjusted to make its adjustment effect better. In some cases, the monitoring system 20 can determine what kind of problem exists based on the monitoring score and the deviation score of each parameter, and give corresponding optimization prompt information based on the corresponding problem. For example, when it is determined that there is an adjustment timeout problem, it can be recommended to perform the following operations in sequence: shorten the integral time, reduce the proportional band, and increase the differential time.
[0149] In this way, the monitoring system 20 can monitor, evaluate and optimize the regulation quality of the control loop, and more intuitively and effectively reflect the regulation quality of the control loop.
[0150] It should be noted that not all steps and modules in the above-mentioned processes and system structure diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The system structure described in the above-mentioned embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or some components in multiple independent devices may be implemented together.
[0151] In the above embodiments, the hardware module can be implemented mechanically or electrically. For example, a hardware module can include a permanent dedicated circuit or logic (such as a dedicated processor, FPGA or ASIC) to complete the corresponding operation. The hardware module can also include programmable logic or circuits (such as a general-purpose processor or other programmable processors), which can be temporarily set by software to complete the corresponding operation. The specific implementation method (mechanical method, or dedicated permanent circuit, or temporarily set circuit) can be determined based on cost and time considerations.
[0152] The present application is presented and described in detail above through the accompanying drawings and preferred embodiments; however, the present application is not limited to these disclosed embodiments, and those skilled in the art may know based on the above-mentioned multiple embodiments that the code review methods in the above-mentioned different embodiments may be combined to obtain more embodiments of the present application, and these embodiments are also within the protection scope of the present application.
Claims
1. A control loop monitoring method, characterized in that: The method is applied to a device configured with a control loop and a controlled object, wherein the control loop includes a controller and an actuator, and the controller controls the actuator to act to adjust the controlled object. The method includes: Determine the start time of the control loop to adjust the regulated object, and obtain the target value and actual value of the regulated parameter of the regulated object and the time when the actual value appears, as well as the control instruction information and feedback information of the actuator; Determining whether the regulated object and the control loop are in a steady state according to an actual value of a regulated parameter of the regulated object and a time when the actual value appears, the control instruction information and the feedback information; If it is in a steady state, the monitoring information of the regulation quality of the control loop is determined based on at least part of the regulation start time, the actual value and the time when the actual value appears, the target value, the control instruction information and the feedback information.
2. The monitoring method according to claim 1, characterized in that: The step of determining the start time of regulating the regulated object by the control loop includes: The time when the target value of the adjusted parameter configured in the control loop changes to a new target value is determined as the adjustment start time, or the time when the actual value of the adjusted parameter of the adjusted object exceeds the set interval is determined as the adjustment start time.
3. The monitoring method according to claim 1, characterized in that: The determining whether the regulated object and the control loop are in a steady state comprises: If the duration of the actual value being in the first steady-state interval satisfies the first duration, it is determined that the regulated object is in a steady state, wherein the first steady-state interval is determined according to the target value and a set first error; or, If the duration of the actual state value of the actuator indicated by the feedback information being in a second steady-state interval satisfies a second duration, it is determined that the control loop is in a steady state, wherein the second steady-state interval is determined according to the target state value and a preset second error.
4. The monitoring method according to any one of claims 1 to 3, characterized in that: The monitoring information for determining the regulation quality of the control loop includes at least one of the following: Determine a steady-state moment when the control loop and the regulated object are in a steady state according to the actual value and the corresponding time, the target value, the control instruction information and the feedback information, and determine the regulation duration of the control loop according to the steady-state moment and the regulation start time; Determine a first adjustment oscillation duration of the regulated object according to the actual value and the corresponding time and the target value, determine a second adjustment oscillation duration of the actuator according to the control instruction information and the feedback information, and determine an oscillation duration of the control loop based on the first adjustment oscillation duration and the second adjustment oscillation duration; Determining a regulation deviation according to a difference between the actual value and the target value when the regulated object and the control loop are in a steady state; When the actual value and the target value overlap, an overshoot value of the control loop is determined according to the actual value and the target value.
5. The monitoring method according to claim 4, characterized in that: The determining of the first adjustment oscillation duration of the adjusted object includes: Determine a first moment when the actual value reaches a first steady-state interval, where the first steady-state interval is determined according to the target value and a set first error; Determine a second time when the duration of the actual value being in the first steady-state interval satisfies a first duration; The time length between the first moment and the second moment is used as the first adjustment oscillation time length.
6. The monitoring method according to claim 5, characterized in that: Determining a second adjustment oscillation duration of the actuator according to the control instruction information and the feedback information includes: Determine a third moment when the duration of the actual state value of the actuator indicated by the feedback information being within a second steady-state interval satisfies the second duration, wherein the second steady-state interval is determined according to the target state value indicated by the control instruction information and a preset second error; If the third moment is earlier than or equal to the second moment, the second adjustment oscillation duration is 0; or, if the third moment is later than the second moment, the second adjustment oscillation duration is the duration between the third moment and the second moment.
7. The monitoring method according to claim 4, characterized in that: The monitoring information for determining the regulation quality of the control loop further includes: The variance of the regulated object and the regulated parameter in the oscillation phase of the control loop is determined according to the actual value and the corresponding time and the target value.
8. The monitoring method according to any one of claims 1 to 3, characterized in that: The method further comprises: Determining at least one of a control time constant of the control loop, a damping time of the regulated object, and a damping time of the actuator; Wherein, the control time constant is the time between the start time of the adjustment and a fourth time, and the fourth time is the time when the actuator starts to act as indicated by the feedback information; The damping time of the regulated object is the time between a first moment and a second moment, the first moment is the moment when the actual value reaches a first steady-state interval, and the second moment is the moment when the actual value is in the first steady-state interval for a duration that satisfies a first duration; The damping time of the actuator is the time between the fifth moment and the third moment, the fifth moment is the moment when the actual state value indicated by the feedback information reaches the second steady-state interval, and the third moment is the moment when the duration of the actual state value in the second steady-state interval satisfies the second duration.
9. The monitoring method according to any one of claims 1 to 3, characterized in that: The method further comprises: Determine, according to at least one of a switch signal of a switch corresponding to the control loop and an actual value of the regulated object, a first working time duration during which the control loop is in a working state for regulating the regulated object and a second working time duration during which the device is in a working state; The utilization rate of the control loop is determined according to the first working duration and the second working duration.
10. A control loop evaluation method, characterized in that: include: Acquire monitoring information for indicating the quality of regulation of the control loop on the regulated object, wherein the monitoring information is acquired by the method according to any one of claims 1 to 9, and the monitoring information includes at least one of the following parameters: regulation duration, oscillation duration, regulation deviation, overshoot value and variance; Acquire benchmark information and weights corresponding to the monitoring information, wherein the benchmark information includes at least one of the following benchmark values: an adjustment duration benchmark value, an oscillation duration benchmark value, an adjustment deviation benchmark value, an overshoot benchmark value, and a variance benchmark value; A monitoring score of the control loop is determined based on the monitoring information, the preset benchmark information and the weight.
11. The evaluation method according to claim 10, characterized in that: Determining the monitoring score of the control loop includes: For each parameter in the monitoring information, determine the deviation score of the parameter according to the difference ratio between the parameter and the corresponding reference value, and calculate the product of the deviation score of the parameter and the corresponding weight as the weighted result of the parameter; The monitoring score is determined according to the sum of the weighted results of each of the parameters.
12. A control loop optimization method, characterized in that: include: Obtaining a monitoring score and a deviation score of the control loop, wherein the monitoring score and the deviation score are obtained by the method of claim 10 or 11; Optimization prompt information is determined based on the monitoring score, the deviation score and preset optimization conditions.
13. The optimization method according to claim 12, characterized in that: The optimization condition includes at least one of the following: an adjustment timeout condition, a deviation condition, and an oscillation condition; and the determination of the optimization prompt information according to the monitoring score, the deviation score, and the preset optimization condition includes at least one of the following: If it is determined that the adjustment timeout condition is met according to the monitoring score and the deviation score, a first optimization prompt information is output, wherein the first optimization prompt information indicates to perform the following operations in sequence: shorten the integral time, reduce the proportional band, and increase the differential time; If it is determined that the deviation condition is met according to the monitoring score and the deviation score, a second optimization prompt information is output, wherein the second optimization prompt information indicates to perform the following operations in sequence: reducing the proportional band, increasing the integral time, and increasing the differential time; If it is determined that the oscillation condition is met according to the monitoring score and the deviation score, a third optimization prompt information is output, and the third optimization prompt information indicates to perform the following operations in sequence: increase the integral time, increase the differential time, and enlarge the proportional band.
14. A control loop monitoring system, characterized in that: The monitoring system is connected to a device equipped with the control loop, and the monitoring system is used to execute the method according to any one of claims 1 to 13.
