A method and device suitable for on-line analysis of process control performance of thermal power plants

Abstract

This invention discloses a method and apparatus for online analysis of process control performance in thermal power plants. The method includes: collecting process operation data from the thermal power plant and performing data preprocessing; determining and recording the moment when the controller of the thermal power plant begins to experience a disturbance and the type of the disturbance; after the disturbance begins, determining whether it is a valid peak / trough; if it is a valid peak / trough, recording the corresponding peak / trough value and the time of its occurrence; if it is an invalid peak / trough, determining whether the disturbance has ended; if the system enters a steady state or oscillates, determining that the disturbance has ended, and resetting the relevant status flags; if the system has not entered a steady state, issuing an alarm and returning to execute the above process. The advantage of this invention is that it can determine whether the controller of a thermal power generating unit is stable and accurately monitor its performance parameters, thereby enabling the controller to adjust to a steady state in a timely manner when data oscillations occur, avoiding impact on actual production.

Description

Technical Field

[0001] This invention relates to the field of thermal automation, and more specifically to a method and apparatus for online analysis of process control performance in thermal power plants. Background Technology

[0002] In the process control of thermal power generating units, the control performance of the controller is reflected in whether the system can respond stably, accurately, and quickly after being disturbed. Because thermal power generating units involve a wide range of control operations, a large amount of process operation data is generated during the control process. The changing trends of this data include information such as overshoot, settling time, and steady-state of the controlled object. By collecting and monitoring this process operation data, the stability of the controller can be determined. Current technologies mainly collect process operation data through field tests and rely on human experience to judge the rationality of the system's control parameters, thereby making corresponding adjustments. However, the effectiveness of parameter adjustments largely depends on the richness of human experience, resulting in a certain degree of randomness in system parameter adjustments.

[0003] Chinese Patent Publication No. CN111123867A discloses a DCS system for thermal power plants, including a process parameter acquisition module for acquiring process control parameters of the thermal power plant; a data storage module for storing process control data; a data integration module for integrating process control parameters; a control command generation module for generating control commands; and an execution module for executing control commands. This patent application effectively reduces the data processing load of the DCS system, but it cannot determine whether the controller of the thermal power generating unit is stable and cannot accurately monitor its performance parameters. This results in the controller being unable to adjust to a steady state in a timely manner when data oscillations occur, affecting actual production. Summary of the Invention

[0004] The technical problem to be solved by this invention is that the existing technology cannot determine whether the controller of a thermal power generating unit is stable and cannot accurately monitor its performance parameters, so that the controller cannot adjust to a steady state in time when data oscillation occurs, which affects actual production.

[0005] This invention solves the above-mentioned technical problems through the following technical means: a method for online analysis of process control performance in thermal power plants, comprising the following steps:

[0006] Step 1: Collect process operation data from the thermal power plant and perform data preprocessing;

[0007] Step 2: Determine and record the moment when the power plant's controller begins to experience a disturbance and the type of disturbance.

[0008] Step 3: After the perturbation starts, determine whether it is a valid peak / trough. If it is a valid peak / trough, record the corresponding peak / trough value and the time when it occurs. If it is an invalid peak / trough, proceed to Step 4;

[0009] Step 4: Determine whether this perturbation has ended. If the system enters a steady state or oscillates, it is determined that this perturbation has ended, and the relevant status flag bits are reset. If the system does not enter a steady state, the system issues an alarm and returns to execute Step 3 and Step 4.

[0010] Beneficial effects: The present invention collects the process operation data and records the time when the perturbation pair starts to appear and the type of perturbation. After the perturbation starts, it determines whether it is a valid peak or trough value. If it is valid, it records the corresponding value and the time when it occurs, so as to accurately record the perturbation time, perturbation type and verify the authenticity of the perturbation, realize the accurate monitoring of performance parameters, and can judge in real time whether the system enters a steady state. Therefore, measures can be taken in time when the system does not enter a steady state or oscillates, ensuring that it quickly enters a steady state and avoiding affecting production.

[0011] Further, the said Step 1 includes: collecting the process operation data of the thermal power plant and filtering the noise signals in the process operation data.

[0012] Further, the said Step 2 includes:

[0013] Step 201: If the absolute value of the deviation between the set value SP and the measured value PV is greater than the PID regulation dead zone err_d, it is determined that the system perturbation starts. Record the set value SP and the measured value PV at the perturbation start time, and proceed to Step 202;

[0014] Step 202: Judge whether the absolute value of the deviation between the set value SP and the measured value PV is greater than the steady-state error e SS , if the judgment result is true, it is determined as a real perturbation and proceed to Step 203, otherwise it is considered as interference;

[0015] Step 203: Judge whether the absolute value of the deviation between the set value SP(n) at the current time and the set value SP at the perturbation start time (n0) is greater than or equal to the PID regulation dead zone err_d. If |SP (n) - SP (n0) |≥ err_d, it is determined as a set value perturbation. If |SP (n) - SP (n0) |<err_d, it is determined as a controlled variable perturbation.

[0016] Further, the determination method of the valid peak / trough in the said Step 3 is:

[0017] Record the peak values and the moments when the peaks occur, and correspondingly form peak arrays and peak time arrays. Record the valley values and the moments when the valleys occur, and correspondingly form valley arrays and valley time arrays. If the peak value at the current moment is greater than the values collected at two adjacent moments, and the deviation between the peak value at the current moment and the previous valley value is greater than the peak-valley dead zone requirement value, and the time interval between the moment when the current valley occurs and the moment when the previous valley occurs is greater than the preset sampling time, then the current peak value is a valid peak value and the current valley value is a valid valley value.

[0018] Further, the method for judging that the system enters a steady state in step four is as follows:

[0019] If |SP(n) - PV(n)| < e SS or |PV(n) - PV(n1)| < e3 and |SP(n) - SP(n1)| < e4, then the system enters a steady state, where SP is the set value, PV is the measured value, e SS is the steady-state error, PV(n) is the measured value at the current moment, PV(n1) is the measured value at the previous moment, e3 represents the dead zone of the measured value change, SP(n) is the set value at the current moment, SP(n1) is the set value at the previous moment, and e4 represents the dead zone of the set value change.

[0020] Furthermore, the method for judging that the system oscillates in step four is as follows:

[0021] When the number of data in both the peak array and the valley array is greater than or equal to 3, if the deviation between the peak value and the valley value at the same index remains unchanged or continuously increases, it is determined that the system oscillates.

[0022] The present invention also provides a device applicable to on-line analysis of the process control performance of a thermal power plant, including:

[0023] A data preprocessing module, which is used to collect the process operation data of the thermal power plant and perform data preprocessing;

[0024] A disturbance judgment module, which is used to judge and record the moment when the controller of the thermal power plant starts to have a disturbance and the type of this disturbance;

[0025] A valid peak-valley value judgment module, which is used to judge whether it is a valid peak / valley after the disturbance starts. If it is a valid peak / valley, record the corresponding peak / valley value and the moment when it occurs. If it is an invalid peak / valley, execute the disturbance end judgment module;

[0026] A disturbance end judgment module, which is used to judge whether this disturbance ends. If the system enters a steady state or oscillates, it is determined that this disturbance ends, and then reset the relevant status flags. If the system does not enter a steady state, the system issues an alarm and returns to execute the valid peak-valley value judgment module and the disturbance end judgment module.

[0027] Further, the data preprocessing module is further configured to: collect the process operation data of the thermal power plant and filter the noise signals in the process operation data.

[0028] Further, the disturbance judgment module is further configured to:

[0029] Step 201: If the absolute value of the deviation between the set value SP and the measured value PV is greater than the PID regulation dead zone err_d, it is determined that the system disturbance starts, record the set value SP and the measured value PV at the disturbance start time, and enter Step 202;

[0030] Step 202: Judge whether the absolute value of the deviation between the set value SP and the measured value PV is greater than the steady-state error e SS , if the judgment result is true, it is determined as a real disturbance, enter Step 203, otherwise it is considered as interference;

[0031] Step 203: Judge whether the absolute value of the deviation between the set value SP(n) at the current time and the set value SP at the disturbance start time (n0) is greater than or equal to the PID regulation dead zone err_d. If |SP (n) - SP (n0) |≥ err_d, it is determined as a set value disturbance. If |SP (n) - SP (n0) |<err_d, it is determined as a controlled variable disturbance.

[0032] Further, the determination method of the effective peak and valley values in the effective peak and valley value judgment module is:

[0033] Record the peak value and the time when the peak occurs, correspondingly form a peak array and a peak time array, record the valley value and the time when the valley occurs, correspondingly form a valley array and a valley time array. If the peak value at the current time is greater than the values collected at the adjacent two times, and the deviation between the peak value at the current time and the previous valley value is greater than the peak-valley dead zone requirement value, and the time interval between the time when the current valley occurs and the time when the previous valley occurs is greater than the preset sampling time, then the current peak value is an effective peak value and the current valley value is an effective valley value.

[0034] Further, the judgment method for the system to enter the steady state in the disturbance end judgment module is:

[0035] If |SP(n)-PV(n)|<e SS or |PV(n)- PV(n1)|<e3 and |SP(n)- SP(n1)|<e4, the system enters the steady state, where SP is the set value, PV is the measured value, and e SSFor steady-state error, PV(n) is the current measured value, PV(n1) is the previous measured value, e3 represents the dead zone of the measured value change, SP(n) is the current set value, SP(n1) is the previous set value, and e4 represents the dead zone of the set value change.

[0036] Furthermore, the method for determining system oscillation in the disturbance termination determination module is as follows:

[0037] When the number of data in both the peak array and the trough array is greater than or equal to 3, if the deviation between the peak value and the trough value at the same index remains unchanged or continues to increase, then the system is determined to be oscillating.

[0038] The advantages of this invention are as follows: This invention collects process operation data and records the time and type of disturbance that begins to appear. After the disturbance begins, it determines whether it is a valid peak or trough value. If it is valid, it records the corresponding value and the time of occurrence. This allows for accurate recording of the disturbance time and type, as well as verification of the disturbance's authenticity. This enables accurate monitoring of performance parameters and allows for real-time determination of whether the system has entered a steady state. This allows for timely measures to be taken if the system has not entered a steady state or if oscillations occur, ensuring that it quickly enters a steady state and avoiding any impact on production. Attached Figure Description

[0039] Figure 1 This is a flowchart of a method for online analysis of process control performance in thermal power plants, as disclosed in Embodiment 1 of the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] Example 1

[0042] like Figure 1 As shown, a method for online analysis of process control performance in thermal power plants includes the following steps:

[0043] S1: Collect process operation data from thermal power plants and filter noise signals in the process operation data;

[0044] S2: Determine and record the moment when the thermal power plant's controller begins to experience a disturbance, as well as the type of this disturbance; the specific process is as follows:

[0045] Step 201: If the absolute value of the deviation between the set value SP and the measured value PV is greater than the PID regulation dead zone err_d, it is determined that the system disturbance starts. Record the set value SP and the measured value PV at the start time of the disturbance, and enter Step 202;

[0046] Step 202: Judge whether the absolute value of the deviation between the set value SP and the measured value PV is greater than the steady-state error e SS , if the judgment result is true, it is determined as a real disturbance, and enter Step 203, otherwise it is considered as interference;

[0047] Step 203: Judge whether the absolute value of the deviation between the set value SP(n) at the current time and the set value SP at the start time of the disturbance (n0) is greater than or equal to the PID regulation dead zone err_d. If |SP (n) - SP (n0) |≥ err_d, it is determined as a set value disturbance. If |SP (n) - SP (n0) |<err_d, it is determined as a regulated variable disturbance.

[0048] S3: After the disturbance starts, judge whether it is a valid peak / trough. If it is a valid peak / trough, record the corresponding peak / trough value and the time when it occurs. If it is an invalid peak / trough, enter S4; Among them, the determination method of the valid peak / trough is:

[0049] Record the peak value and the time when the peak occurs, and correspondingly form a peak array and a peak time array. Record the trough value and the time when the trough occurs, and correspondingly form a trough array and a trough time array. If the peak value at the current time is greater than the values collected at the two adjacent times, and the deviation between the peak value at the current time and the previous trough value is greater than the peak-trough dead zone requirement value, and the time interval between the time when the current trough occurs and the time when the previous trough occurs is greater than the preset sampling time, then the current peak value is a valid peak value, and the current trough value is a valid trough value. By this way, the influence of random disturbances can be effectively avoided while locking the peak.

[0050] S4: Judge whether this disturbance ends. If the system enters the steady state or oscillates, it is determined that this disturbance ends, and then reset the relevant status flag bits. If the system does not enter the steady state, the system issues an alarm and returns to execute S3 and S4.

[0051] Among them, the judgment method for the system to enter the steady state is:

[0052] If |SP(n)-PV(n)|<e SSIf |PV(n) - PV(n1)| < e3 and |SP(n) - SP(n1)| < e4, the system enters the steady state, where SP is the set value, PV is the measured value, e SS is the steady-state error, PV(n) is the measured value at the current moment, PV(n1) is the measured value at the previous moment, e3 represents the dead zone of the measured value change, SP(n) is the set value at the current moment, SP(n1) is the set value at the previous moment, and e4 represents the dead zone of the set value change.

[0053] Among them, the method for judging the oscillation of the system is as follows:

[0054] When the number of data in both the peak array and the valley array is greater than or equal to 3, if the deviation between the peak value and the valley value at the same index remains unchanged or continuously increases, it is determined that the system oscillates. The peak array and the valley array take values in the array with the same subscript, and the deviation between the peak value and the valley value is obtained. If the deviation value remains unchanged or continuously increases, it is determined that the system oscillates. For example: Let the peak array be A and the valley array be B, a1 = A[1] - B[1]; a2 = A[2] - B[2]; a3 = A[3] - B[3]. If a3 = a2 = a1 or a3 > a2 > a1, it means the system oscillates.

[0055] Through the above technical solutions, the present invention collects the process operation data during the collection process and records the moment when the disturbance pair starts to appear and the type of disturbance, and judges whether it is an effective peak-valley value after the disturbance starts. If it is effective, the corresponding value and the occurrence moment are recorded, so as to accurately record the disturbance moment, the disturbance type and verify the authenticity of the disturbance, realize the accurate monitoring of the performance parameters, and can judge in real time whether the system enters the steady state, so as to take measures in time when the system does not enter the steady state or oscillates, ensure its rapid entry into the steady state, and avoid affecting the production.

[0056] Embodiment 2

[0057] Based on Embodiment 1, Embodiment 2 of the present invention further provides a device suitable for on-line analysis of the process control performance of a thermal power plant, including:

[0058] A data preprocessing module, which is used to collect the process operation data of the thermal power plant and perform data preprocessing;

[0059] A disturbance judgment module, which is used to judge and record the moment when the controller of the thermal power plant starts to appear disturbed and the type of this disturbance;

[0060] An effective peak-valley value judgment module, which is used to judge whether it is an effective peak / valley after the disturbance starts. If it is an effective peak / valley, the corresponding peak / valley value and the occurrence moment are recorded. If it is an invalid peak / valley, the disturbance end judgment module is executed;

[0061] The disturbance end judgment module is used to judge whether the current disturbance ends. If the system enters a steady state or oscillates, it is determined that the current disturbance ends, and the relevant status flags are reset. If the system does not enter a steady state, the system issues an alarm and returns to execute the effective peak-valley value judgment module and the disturbance end judgment module.

[0062] Specifically, the data preprocessing module is also used to: collect the process operation data of the thermal power plant and filter the noise signals in the process operation data.

[0063] Specifically, the disturbance judgment module is also used to:

[0064] Step 201: If the absolute value of the deviation between the set value SP and the measured value PV is greater than the PID regulation dead zone err_d, it is determined that the system disturbance starts. Record the set value SP and the measured value PV at the moment when the disturbance starts, and enter Step 202;

[0065] Step 202: Judge whether the absolute value of the deviation between the set value SP and the measured value PV is greater than the steady-state error e SS , if the judgment result is true, it is determined as a real disturbance and enter Step 203, otherwise it is considered as interference;

[0066] Step 203: Judge whether the absolute value of the deviation between the set value SP(n) at the current moment and the set value SP at the disturbance start moment (n0) is greater than or equal to the PID regulation dead zone err_d. If |SP (n) - SP (n0) |≥ err_d, it is determined as a set value disturbance. If |SP (n) - SP (n0) |<err_d, it is determined as a controlled variable disturbance.

[0067] Specifically, the determination method of the effective peak and valley in the effective peak-valley value judgment module is as follows:

[0068] Record the peak value and the moment when the peak occurs, and correspondingly form a peak array and a peak time array. Record the valley value and the moment when the valley occurs, and correspondingly form a valley array and a valley time array. If the peak value at the current moment is greater than the values collected at the adjacent two moments, and the deviation between the peak value at the current moment and the previous valley value is greater than the peak-valley dead zone requirement value, and the time interval between the moment when the current valley occurs and the moment when the previous valley occurs is greater than the preset sampling time, then the current peak value is an effective peak value and the current valley value is an effective valley value.

[0069] Specifically, the judgment method for the system to enter a steady state in the disturbance end judgment module is as follows:

[0070] If |SP(n) - PV(n)| < e SS or |PV(n) - PV(n1)| < e3 and |SP(n) - SP(n1)| < e4, the system enters the steady state, where SP is the set value, PV is the measured value, e SS is the steady-state error, PV(n) is the measured value at the current moment, PV(n1) is the measured value at the previous moment, e3 represents the dead zone of the measured value change, SP(n) is the set value at the current moment, SP(n1) is the set value at the previous moment, and e4 represents the dead zone of the set value change.

[0071] More specifically, the method for judging the oscillation of the system in the disturbance end judgment module is as follows:

[0072] When the number of data in both the peak array and the valley array is greater than or equal to 3, if the deviation between the peak value and the valley value at the same index remains unchanged or continuously increases, it is determined that the system oscillates.

[0073] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for online performance analysis of process control in thermal power plants, characterized in that, Includes the following steps: Step 1: Collect process operation data from the thermal power plant and perform data preprocessing; Step 2: Determine and record the moment when the power plant's controller begins to experience a disturbance and the type of disturbance. Step 3: After the disturbance begins, determine whether it is a valid peak / trough. If it is a valid peak / trough, record the corresponding peak / trough value and the time when it occurred. If it is an invalid peak / trough, proceed to step 4. Step 4: Determine whether the disturbance has ended. If the system enters a steady state or oscillates, the disturbance is considered to have ended, and the relevant status flags are reset. If the system has not entered a steady state, the system issues an alarm and returns to execute steps 3 and 4.

2. The method for online performance analysis of process control in thermal power plants according to claim 1, characterized in that, Step one includes: collecting process operation data of the thermal power plant and filtering noise signals in the process operation data.

3. The method for online performance analysis of process control in thermal power plants according to claim 1, characterized in that, Step two includes: Step 201: If the absolute value of the deviation between the setpoint SP and the measured value PV is greater than the PID control dead zone err_d, then the system disturbance is determined to have started. Record the setpoint SP and the measured value PV at the time the disturbance starts, and proceed to step 202. Step 202: Determine whether the absolute value of the deviation between the setpoint SP and the measured value PV is greater than the steady-state error e. SS If the result is true, it is determined to be a real disturbance and proceeds to step 203; otherwise, it is considered to be interference. Step 203: Determine the current time setting value SP(n) and the disturbance start time setting value SP. (n0) If the absolute value of the deviation between them is greater than or equal to the PID control dead zone err_d, then |SP (n) - SP (n0) If |≥ err_d, it is determined to be a setpoint disturbance; if |SP (n) - SP (n0) If |< err_d, it is determined to be a disturbance of the controlled variable.

4. The method for online performance analysis of process control in thermal power plants according to claim 1, characterized in that, The method for determining the effective peaks / troughs in step three is as follows: Record the peak value and the time of the peak occurrence, forming a peak array and a peak time array. Record the trough value and the time of the trough occurrence, forming a trough array and a trough time array. If the peak value at the current time is greater than the values ​​collected at the two adjacent times, and the deviation between the peak value at the current time and the previous trough value is greater than the peak-trough dead zone requirement, and the time interval between the time of the current trough occurrence and the time of the previous trough occurrence is greater than the preset sampling time, then the current peak value is a valid peak value, and the current trough value is a valid trough value.

5. The method for online performance analysis of process control in thermal power plants according to claim 1, characterized in that, The method for determining whether the system has entered a steady state in step four is as follows: If |SP(n)-PV(n)|< e SS If |PV(n) - PV(n1)| < e3 and |SP(n) - SP(n1)| < e4, then the system enters steady state, where SP is the setpoint, PV is the measured value, and e4 is the measured value. SS For steady-state error, PV(n) is the current measured value, PV(n1) is the previous measured value, e3 represents the dead zone of the measured value change, SP(n) is the current set value, SP(n1) is the previous set value, and e4 represents the dead zone of the set value change.

6. The method for online performance analysis of process control in thermal power plants according to claim 4, characterized in that, The method for determining system oscillation in step four is as follows: When the number of data in both the peak array and the trough array is greater than or equal to 3, if the deviation between the peak value and the trough value at the same index remains unchanged or continues to increase, then the system is determined to be oscillating.

7. A device suitable for online analysis of process control performance in thermal power plants, characterized in that, include: The data preprocessing module is used to collect process operation data from thermal power plants and perform data preprocessing. The disturbance detection module is used to determine and record the moment when the controller of the thermal power plant begins to experience a disturbance and the type of the disturbance. The effective peak / valley value judgment module is used to determine whether a peak / valley is valid after the disturbance begins. If it is a valid peak / valley, the corresponding peak / valley value and the time of its occurrence are recorded. If it is an invalid peak / valley, the disturbance end judgment module is executed. The disturbance termination judgment module is used to determine whether the disturbance has ended. If the system enters a steady state or oscillates, the disturbance is determined to have ended, and the relevant status flags are reset. If the system has not entered a steady state, the system issues an alarm and returns to execute the valid peak and valley value judgment module and the disturbance termination judgment module.

8. The device for online analysis of process control performance in thermal power plants according to claim 7, characterized in that, The data preprocessing module is also used to: collect process operation data of thermal power plants and filter noise signals in the process operation data.

9. The device for online performance analysis of process control in thermal power plants according to claim 7, characterized in that, The disturbance detection module is also used for: Step 201: If the absolute value of the deviation between the setpoint SP and the measured value PV is greater than the PID control dead zone err_d, then the system disturbance is determined to have started. Record the setpoint SP and the measured value PV at the time the disturbance starts, and proceed to step 202. Step 202: Determine whether the absolute value of the deviation between the setpoint SP and the measured value PV is greater than the steady-state error e. SS If the result is true, it is determined to be a real disturbance and proceeds to step 203; otherwise, it is considered to be interference. Step 203: Determine the current time setting value SP(n) and the disturbance start time setting value SP. (n0) If the absolute value of the deviation between them is greater than or equal to the PID control dead zone err_d, then |SP (n) - SP (n0) If |≥ err_d, it is determined to be a setpoint disturbance; if |SP (n) - SP (n0) If |< err_d, it is determined to be a disturbance of the controlled variable.

10. The device for online analysis of process control performance in thermal power plants according to claim 7, characterized in that, The method for determining valid peaks / valleys in the valid peak / valley value determination module is as follows: Record the peak value and the time of the peak occurrence, forming a peak array and a peak time array. Record the trough value and the time of the trough occurrence, forming a trough array and a trough time array. If the peak value at the current time is greater than the values ​​collected at the two adjacent times, and the deviation between the peak value at the current time and the previous trough value is greater than the peak-trough dead zone requirement, and the time interval between the time of the current trough occurrence and the time of the previous trough occurrence is greater than the preset sampling time, then the current peak value is a valid peak value, and the current trough value is a valid trough value.

Images