Fault diagnosis method and device for control circuit of refinery

Abstract

An embodiment of the present invention provides a fault diagnosis method for a control loop of a refinery device, comprising: obtaining historical data parameters of one or more of the following in the control loop: a control valve, auxiliary equipment, a measuring instrument, and a PID parameter model, the historical data parameters comprising: a set value SP, a measured value PV, and an output value MV of the control loop; calculating state information of the control loop according to the acquired historical data parameters; and determining the cause of the fault of the control loop according to the state information, thereby improving the self-control rate and stability of the device and reducing the probability of unplanned shutdown.

Description

Technical Field

[0001] The present invention relates to the technical field of control loop diagnosis, and in particular to a fault diagnosis method and device for a control loop of a refinery. Background Art

[0002] The operating performance of the control loop has a huge impact on production stability, production efficiency, and even production safety. If there is a disturbance in our production process, the most effective and safest way is to automatically adjust it through the conventional control loop. In this way, the user's loss will be minimized. If the conventional loop adjustment fails, an alarm will be issued and the operator will intervene manually. If the intervention is unsuccessful, a safety interlock will be performed, and a safe stop will be performed. The loss and danger will increase with each increase in the degree of intervention in the handling of these production disturbances. The most ideal state is to eliminate the disturbance through the automatic adjustment of the conventional control loop. Therefore, the performance of the control loop becomes very important and is directly related to production stability, efficiency and production safety.

[0003] Due to factors such as control valve stagnation, sensor unit failure, frequent external disturbances, unreasonable control strategy, unreasonable PID parameter settings, poor process and poor process control design, the control loop performance is poor. In severe cases, it can only be controlled manually, affecting the automatic control utilization rate of the device. Poor control loop performance brings the following adverse consequences: causing frequent operation of the control valve, increasing power drive costs, reducing service life, and increasing maintenance costs; the device's anti-interference ability is insufficient, and it cannot adjust abnormal working conditions in time, resulting in a large number of invalid alarms due to large fluctuations, affecting the stability of the device; affecting the device's edge operation and generation, failing to achieve the optimal production environment, and not conducive to refined management. Summary of the invention

[0004] The purpose of the embodiment of the present invention is to provide a fault diagnosis method for a control loop of a refinery, which calculates the state information of the control loop according to the set value SP, the measured value PV and the output value MV of the control loop, and then determines the cause of the fault of the control loop according to the calculated state information. According to the commissioning and performance status of the control loop of the refinery, through the current status analysis and performance evaluation of the control loop, control valve fault diagnosis, instrument failure diagnosis, PID parameter model fault diagnosis and / or PID parameter tuning, the control loop and control valve of the device are analyzed, optimized, monitored and managed, the problem control loop is reduced, the automatic control rate and stability of the device are improved, and the probability of unplanned shutdown is reduced.

[0005] In order to achieve the above object, an embodiment of the present invention provides a fault diagnosis method for a control loop of a refinery, the method comprising:

[0006] Obtaining historical data parameters of one or more of the following in the control loop: a control valve, an auxiliary device, a measuring instrument, and a PID parameter model, wherein the historical data parameters include: a set value SP, a measured value PV, and an output value MV of the control loop;

[0007] Calculating the state information of the control loop according to the acquired historical data parameters; and

[0008] A fault cause of the control loop is determined according to the status information.

[0009] Optionally, the status information includes:

[0010] A linear relationship between the opening setting value SP of the control valve and the actual output value MV of the controlled variable associated with the control valve;

[0011] The deviation unilateral rate and output saturation rate related to the measured value PV, the set value SP and the output value MV of the control loop;

[0012] The similarity between the measured value PV of the control loop and the measured estimate; and / or

[0013] Deviation time series data of the control loop.

[0014] Optionally, the linear relationship between the opening setting value SP of the control valve and the actual output value MV of the controlled variable related to the control valve is calculated according to the following formula:

[0015]

[0016] Among them, ρ (n) represents the sequence of correlation coefficients between the opening setting value of the control valve and the actual output value of the controlled variable associated with the control valve, PV (i) represents the measured value at the i-th moment, MV (i) Represents the valve position value at the i-th moment, Index is |ρ (n) |The location of the maximum value.

[0017] Among them, in the Index and ρ (Index) When the coefficient is within the preset range, it is determined that the control valve is in a normal working state; when the Index and ρ (Index) When the preset coefficient range is exceeded, it is determined that the control valve is faulty.

[0018] Optionally, the deviation unilateral rate and the output saturation rate related to the measured value PV, the set value SP and the output value MV of the control loop are calculated according to the following formula:

[0019]

[0020]

[0021] Deviation unilateral rate = max (deviation positive rate, deviation negative rate)

[0022]

[0023] Among them, e is the deviation value between the measured value PV and the set value SP of the auxiliary equipment. The deviation value e has a positive value or a negative value, which causes the output value MV of the auxiliary equipment to have a maximum value and a minimum value. When the deviation unilateral rate is greater than the set unilateral rate and the output saturation rate is greater than the set saturation rate, it is judged that the auxiliary equipment has a fault.

[0024] Optionally, the auxiliary equipment includes one or more of the following: a heat exchanger, an air condenser and an electric heater.

[0025] Optionally, determining the similarity between the measured value PV of the control loop and the estimation of the measured value PV according to a preset NARX neural network instrument estimation model;

[0026] The output value MV of the control loop is used as the NARX neural network instrument valuation model, the output value MV of the control loop and the predicted control loop correction factor α are used as input variables of the NARX neural network instrument valuation model, the measured value PV is used as the output variable to be estimated, and the output value MV and the correction factor α are input into the NARX neural network instrument valuation model to obtain the estimated value PVg of the measured value PV;

[0027] When the similarity between the estimated value PVg and the measured value PV is less than a set similarity, it is determined that the measuring instrument has a fault.

[0028] Optionally, the similarity between the measured value PV and the estimated value PVg is calculated according to the following formula:

[0029]

[0030] Among them, PV i Represents the measured value at time i in the time series, PVg i It represents the predicted value at time i in the time series, and n represents the number of moments in the time series.

[0031] Optionally, the deviation time series data of the control loop is calculated according to the following formula:

[0032] e i =PV i -SP i

[0033]

[0034]

[0035] Among them, PV i is the measured value at time i in the time series, SP i is the set value at time i in the time series, e i PV is the measured value at time i in the time series i With the set value SP i The deviation between i-j PV is the measured value PV of the time series at time i delayed by j time intervals i With the set value SP i The deviation between j is the autocorrelation coefficient of the deviation time series with itself and the time intervals delayed by j, and θ is the arithmetic mean of the autocorrelation coefficients delayed by 1 to m time intervals;

[0036] Among them, on the basis of eliminating control valve, equipment and instrument failures, when the θ value is close to 0, it is judged that the PID parameter model is well set; when the θ value is close to 1, it is judged that there is a problem with the PID parameter model setting, and parameter optimization or model adjustment is required.

[0037] The embodiment of the present invention further provides a fault diagnosis device for a control loop of a refinery, the device comprising:

[0038] A data acquisition unit, used to acquire historical data parameters of one or more of the following in the control loop: a control valve, an auxiliary device, a measuring instrument, and a PID parameter model, wherein the historical data parameters include: a set value SP, a measured value PV, and an output value MV of the control loop;

[0039] A fault diagnosis unit is used to perform the following operations according to the historical data parameters acquired by the data acquisition unit:

[0040] calculating state information of the control loop; and

[0041] A fault cause of the control loop is determined according to the status information.

[0042] Optionally, the status information includes:

[0043] A linear relationship between the opening setting value SP of the control valve and the actual output value MV of the controlled variable associated with the control valve;

[0044] The deviation unilateral rate and output saturation rate related to the measured value PV, the set value SP and the output value MV of the control loop;

[0045] The similarity between the measured value PV of the control loop and the measured estimate; and / or

[0046] Deviation time series data of the control loop.

[0047] Optionally, the linear relationship between the opening setting value SP of the control valve and the actual output value MV of the controlled variable related to the control valve is calculated according to the following formula:

[0048]

[0049] Among them, ρ (n) represents the sequence of correlation coefficients between the opening setting value of the control valve and the actual output value of the controlled variable associated with the control valve, PV (i) represents the measured value at the i-th moment, MV (i) Represents the valve position value at the i-th moment, Index is |ρ (n) |The location of the maximum value.

[0050] Among them, in the Index and ρ (Index) When the coefficient is within the preset range, it is determined that the control valve is in a normal working state; when the Index and ρ (Index) When the preset coefficient range is exceeded, it is determined that the control valve is faulty.

[0051] Optionally, the deviation unilateral rate and the output saturation rate related to the measured value PV, the set value SP and the output value MV of the control loop are calculated according to the following formula:

[0052]

[0053]

[0054] Deviation unilateral rate = max (deviation positive rate, deviation negative rate)

[0055]

[0056] Among them, e is the deviation value between the measured value PV and the set value SP of the auxiliary equipment. The deviation value e has a positive value or a negative value, which causes the output value MV of the auxiliary equipment to have a maximum value and a minimum value. When the deviation unilateral rate is greater than the set unilateral rate and the output saturation rate is greater than the set saturation rate, it is judged that the auxiliary equipment has a fault.

[0057] Optionally, the auxiliary equipment includes one or more of the following: a heat exchanger, an air condenser and an electric heater.

[0058] Optionally, determining the similarity between the measured value PV of the control loop and the estimation of the measured value PV according to a preset NARX neural network instrument estimation model;

[0059] The output value MV of the control loop is used as the NARX neural network instrument valuation model, the output value MV of the control loop and the predicted control loop correction factor α are used as input variables of the NARX neural network instrument valuation model, the measured value PV is used as the output variable to be estimated, and the output value MV and the correction factor α are input into the NARX neural network instrument valuation model to obtain the estimated value PVg of the measured value PV;

[0060] When the similarity between the estimated value PVg and the measured value PV is less than a set similarity, it is determined that the measuring instrument has a fault.

[0061] Optionally, the similarity between the measured value PV and the estimated value PVg is calculated according to the following formula:

[0062]

[0063] Among them, PV i Represents the measured value at time i in the time series, PVg i means . . . , n means . . .

[0064] Optionally, the deviation time series data of the control loop is calculated according to the following formula:

[0065] e i =PV i -SP i

[0066]

[0067]

[0068] Among them, PV i is the measured value at time i in the time series, SPi is the set value at time i in the time series, e i PV is the measured value at time i in the time series i With the set value SP i The deviation between i-j PV is the measured value PV of the time series at time i delayed by j time intervals i With the set value SP i The deviation between j is the autocorrelation coefficient of the deviation time series with itself and the time intervals delayed by j, and θ is the arithmetic mean of the autocorrelation coefficients delayed by 1 to m time intervals;

[0069] Among them, on the basis of eliminating control valve, equipment and instrument failures, when the θ value is close to 0, it is judged that the PID parameter model is well set; when the θ value is close to 1, it is judged that there is a problem with the PID parameter model setting, and parameter optimization or model adjustment is required.

[0070] Through the above technical solution, the state information of the control loop is calculated according to the set value SP, the measured value PV and the output value MV of the control loop, and then the cause of the control loop failure is determined according to the calculated state information. According to the commissioning and performance status of the control loop of the refining unit, through the current status analysis and performance evaluation of the control loop, control valve fault diagnosis, instrument failure diagnosis, PID parameter model fault diagnosis and / or PID parameter tuning, the control loop and control valve of the unit are analyzed, optimized, monitored and managed, the problem control loop is reduced, the automatic control rate and stability of the unit are improved, and the probability of unplanned shutdown is reduced.

[0071] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0073] Figure 1 It is a flow chart of a fault diagnosis method for a control loop of a refinery device provided by an embodiment of the present invention;

[0074] Figure 2 This is a comparison diagram of the control loop valve setting opening and the controlled variable output value provided in the first embodiment of the present invention;

[0075] Figure 3 This is a comparison diagram of the control loop valve setting opening and the controlled variable output value provided in the first embodiment of the present invention;

[0076] Figure 4 is a comparison diagram of the set value SP and the output value MV provided in the fourth embodiment of the present invention;

[0077] Figure 5 It is a structural schematic diagram of a fault diagnosis device for a control loop of a refinery provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0078] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0079] Most of the control systems of refining and chemical plants are based on PID control loops, which are mainly divided into single loops, cascade loops, proportional control loops, split-range control loops, etc. The control loop system mainly includes: measuring instruments, PID parameter models, regulating valves and other auxiliary equipment. The PID parameter model performs relevant algorithms based on the set values ​​and the measured values ​​measured by the measuring instruments, and controls the regulating valves and other auxiliary equipment with given output values ​​to stabilize the measured values ​​at the set values. Therefore, the embodiments of the invention identify the cause of the fault through the main components of the control loop, such as the control valve, auxiliary equipment, measuring instruments, PID parameter models, etc.

[0080] Figure 1 The schematic diagram of the flow chart of the fault diagnosis method of the control loop of the refinery device provided by the embodiment of the present invention is shown as follows: Figure 1 As shown, a fault diagnosis method for a control loop of a refinery, the method comprising: obtaining historical data parameters of one or more of the following in the control loop: a control valve, an auxiliary device, a measuring instrument, and a PID parameter model, the historical data parameters comprising: a set value SP, a measured value PV, and an output value MV of the control loop;

[0081] The following status information of the control loop is calculated based on the acquired historical data parameters:

[0082] A linear relationship between the opening setting value SP of the control valve and the actual output value MV of the controlled variable associated with the control valve;

[0083] The deviation unilateral rate and output saturation rate related to the measured value PV, the set value SP and the output value MV of the control loop;

[0084] The similarity between the measured value PV of the control loop and the measured estimate; and / or

[0085] Deviation time series data of the control loop; and

[0086] A fault cause of the control loop is determined according to the status information.

[0087] Embodiment 1

[0088] Figure 2 FIG. 4 shows a comparison diagram of the control loop valve setting opening and the controlled variable output value provided by the first embodiment of the present invention, such as Figure 2 As shown in the figure, the OPC server DA interface can be used to collect the historical operation data of each control loop of the refinery for one day, including the set value SP of the controlled variable, the actual output value PV of the controlled variable, the opening MV of the control valve, and the manual or automatic state value (MODE) of the control loop during the detection time, etc.

[0089] In a control loop with poor performance, the linear correlation between the opening of the control valve and the actual operating value of the controlled variable is poor. The present invention selects several typical faults that often occur in control valves for example analysis.

[0090] (1) Taking the control valve in the 04FIC01013 control loop as an example, analyze the failure of the control valve and propose corresponding solutions.

[0091] Depend on Figure 2 It can be seen that when the set opening of the control valve becomes larger, the output value of the controlled variable does not change in time, which is caused by the fact that the actual opening of the control valve has not changed; after a period of reaction time, the flow of the conveying medium begins to increase, at this time, the actual opening of the valve begins to increase, and then the controlled variable reaches the maximum value; when the set opening of the control valve becomes smaller, the flow of the conveying medium also begins to decrease after a period of reaction time. This is a typical fault that often occurs in the control valve, the hysteresis phenomenon of the control valve.

[0092] The reason for the hysteresis of the control valve is that during the industrial operation process, the friction, pressure difference and other resistances in the control loop system hinder the opening and closing degree of the control valve, resulting in that when the input signal changes, the valve stem still cannot move and the actual opening cannot be changed.

[0093] (2) Analyze the failure of the control valve using control loop 04PRCA02001 as an example

[0094] Figure 3 The figure shows a comparison diagram between the control loop valve setting opening and the controlled variable output value provided by the first embodiment of the present invention, Figure 3 It can be seen that when the set opening of the control valve increases, the actual pressure value does not change in time, which is different from Figure 2 The reaction is the same, the control valve has hysteresis. The difference is that the set opening of the control valve has changed twice by a small amount, but the actual pressure has not changed. After the set opening has changed by a large amount, the actual pressure suddenly increases. This is another typical fault of the control valve.

[0095] For the fault diagnosis of control valves in control loops, there is currently a practice of combining the set opening of the control valve with the output value of the controlled variable for detailed analysis. Considering that common faults of control valves include hysteresis, jamming, etc., the common point of these faults is that they cause nonlinear characteristics between the set opening of the control valve and the actual output value of the controlled variable. Therefore, the linear correlation between the set opening of the control valve and the actual output value of the controlled variable can be used to determine whether the control valve is faulty. Therefore, the Pearson correlation coefficient between the set opening of the control valve and the actual output value of the measured variable is used to detect whether the control valve is faulty. The calculation formula is as follows:

[0096]

[0097] Index=MATCH(Max(|ρ (n) |))

[0098] Calculated |ρ (Index) |The correlation coefficient value is between 0 and 1. When the linear correlation coefficient is between 0.8 and 1, the linear correlation between the control valve opening and the actual operating value of the controlled variable is good, and the control valve is in normal working condition; when the Pearson correlation coefficient is less than 0.8, the linear correlation between the control valve opening and the actual operating value of the detected variable is poor, and the control valve is very likely to be faulty.

[0099] The calculated Index correlation coefficient value is between 1 and N. When the coordinate of the maximum correlation coefficient is 1 or 2, there is no viscous hysteresis between the opening of the control valve and the actual operating value of the controlled variable, and the control valve is in normal working condition; when the coordinate of the maximum correlation coefficient is greater than 2, there is a viscous hysteresis between the opening of the control valve and the actual operating value of the detected variable. The larger the Index, the more serious the hysteresis, and the control valve is very likely to be faulty.

[0100] Embodiment 2

[0101] 1440 pieces of data of SP value, PV value and MV value of the temperature control loop controlled by a heat exchanger are collected. The SP value is 45, the PV value is less than 45, and the MV value is 100. The deviation unilateral rate is 100%, and the output saturation rate is 100%. This indicates that the heat exchange effect of the heat exchanger is not good, and there may be faults such as scaling and internal leakage, and equipment fault detection and maintenance are required.

[0102] The auxiliary equipment of the control loop is the output control mechanism other than the control valve, mainly including heat exchangers, air coolers, electric heaters, etc. The main causes of failure are poor control effect, failure to open, etc. The failure is manifested in that the measured value does not reach the set value, that is, the deviation e is consistently positive or negative, and the output value MV reaches the maximum or minimum value.

[0103]

[0104]

[0105] Deviation unilateral rate = max (deviation positive rate, deviation negative rate)

[0106]

[0107] When the deviation unilateral rate is too large (greater than 90%) and the output saturation rate is too large (greater than 90%), it indicates that the auxiliary setting control effect is poor and troubleshooting and equipment maintenance are required.

[0108] Embodiment 3

[0109] The NARX neural network instrument valuation model is built using the MATLAB tool for the normal historical operation data of the PV value, MV value and correction factor α of each control loop. The currently collected control loop related parameter operation values ​​are input into the model to obtain the instrument valuation of the measured value. The similarity between the instrument valuation and the measured value is calculated. When the similarity is less than 0.8, it is considered that the instrument is faulty and instrument fault detection and maintenance are required.

[0110] Specifically, the types of instruments in chemical plants mainly include flow, liquid level, pressure, temperature, etc. The causes of instrument failure are aging of components, environmental changes in the measuring part (for example: material adhesion to the pressure probe, stagnation of the float of the liquid level gauge, blockage of the pressure pipe, etc.). The manifestation of instrument failure is generally instrument failure, that is, the measured value and the actual value are greatly different. Since the actual value is reflected by the measured value, it is impossible to obtain real data, and it is necessary to diagnose the instrument failure based on the operating conditions of other parameters. When the external environment such as the load of the device and the upstream of the control loop does not change much, the output value of the control valve is positively correlated with the measurement.

[0111] The upstream pressure and liquid level of the control loop are used as correction factors. The historical operating values ​​of the PV value, MV value and correction factor α of the control loop under normal conditions of the instrument over a period of time are collected, and a neural network instrument valuation model based on NARX is established. The MV and α values ​​are input variables, and the PV value is the output variable to be estimated. The MV and α values ​​collected within the current time range are input into the model to obtain the estimated PV value (PVg). When the PVg value is less similar to the PV value (<80%), the instrument is considered to be malfunctioning.

[0112]

[0113] Among them, the NARX neural network instrument valuation model is a dynamic neural network model based on time series. It uses the output of the neural network as external feedback on the basis of the BP neural network, introduces the input of the neural network, and adds s unit delays to the input. Compared with the static neural network, its output considers the output and input from ns to n at the same time, so it has better performance in the long-term multi-step prediction of nonlinear models.

[0114] Yi=sigmoid(WiXi+bi)

[0115] X1=[x(n-1);…;x(ns);y(n-1);…;y(ns)]

[0116] W1=[WX,WY]

[0117] Yi=Xi+1

[0118] yp(n)=fNARX(x(n),y(n))=Yk

[0119] Wi represents the weight matrix of the i-th layer; bi represents the domain matrix of the i-th layer; Ti represents the conversion function of the i-th layer, which can be a sigmoid function; 1:3 and 1:2 represent the delays of x(t) and y(t) respectively; x(t) is the input of the neural network at time t; y(t) is the actual output at time t; yp(t) is the output of the neural network at time t; x(n) is input into the network and the ones participating in the calculation are x(n-1),…,x(n-3).

[0120] By returning the prediction result along the dotted line as a substitute for the input y(n), a closed-loop NARX neural network can be obtained, which is expressed as:

[0121] yp(n)=fc(x(n))

[0122] On the basis of the closed-loop network, changing the delay of DX to 0:s-1 can obtain a one-step-ahead prediction closed-loop network, whose expression is:

[0123] yp(n+1)=fr_c(x(n))

[0124] By modifying the delay based on the NARX neural network alone, a one-step-ahead prediction network can be obtained, which is expressed as formula (2.10):

[0125] yp(n+1)=fr(x(n),y(n))

[0126] Embodiment 4

[0127] According to the time series data of PV value and SP value of each control loop, the deviation time series data of the control loop is obtained. The autocorrelation coefficient of the delay time interval of 1 to 10 is calculated. When the autocorrelation coefficient is greater than 0.2, it indicates that the PID parameter model setting is unreasonable and the PID parameters or control scheme need to be modified. Figure 4 FIG. 4 shows a comparison diagram of the set value SP and the output value MV provided in the fourth embodiment of the present invention, as shown in FIG. Figure 4 As shown, the deviation of the control loop 04TRC02020 has large periodic changes, so the PID parameter model of this loop is unreasonable.

[0128] Specifically, according to the PID control principle, the ideal control effect of the measured value PV should be equal to the set value SP. Due to measurement errors, noise and other reasons, the PV value has a deviation e relative to the SP value. When the deviation e is a random variable, it proves that the control model is good; when the deviation e has a regularity, it indicates that the PID parameter model is set unreasonably, and the disturbance at the input end cannot be effectively eliminated, and there are oscillations or delays. Establishing an autocorrelation analysis method based on deviation can effectively diagnose the control effect of the current control loop and identify the rationality of the PID parameters.

[0129] e i =PV i -SP i

[0130]

[0131]

[0132] Among them, PV i is the measured value at time i in the time series, SPi is the set value at time i in the time series, e i PV is the measured value at time i in the time series i With the set value SP i The deviation between i-j PV is the measured value PV of the time series at time i delayed by j time intervals i With the set value SP i The deviation between j is the autocorrelation coefficient of the deviation time series with itself and the time intervals delayed by j. θ is the arithmetic mean of the autocorrelation coefficients of the time intervals delayed by 1 to m.

[0133] On the basis of eliminating control valve, equipment and instrument faults, when the θ value is close to 0, the PID parameter model is well set; when the θ value is close to 1, the PID parameter model is poorly set and parameter optimization or model adjustment is required.

[0134] Embodiment 5

[0135] Obtain the set value SP, measured value PV and output value MV of the control loop in the control loop; perform fault diagnosis on the control valve, auxiliary equipment, measuring instrument and PID parameter model according to the above diagnosis method. The above diagnosis method will not be repeated.

[0136] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0137] Figure 5 The structure diagram of the fault diagnosis device of the control loop of the refinery device provided by the embodiment of the present invention is shown as follows: Figure 5 As shown, an embodiment of the present invention further provides a fault diagnosis device for a control loop of a refinery, the device comprising:

[0138] The data acquisition unit 10 is used to acquire historical data parameters of one or more of the following in the control loop: a control valve, an auxiliary device, a measuring instrument, and a PID parameter model, wherein the historical data parameters include: a set value SP, a measured value PV, and an output value MV of the control loop;

[0139] The fault diagnosis unit 20 is configured to perform the following operations according to the historical data parameters acquired by the data acquisition unit:

[0140] calculating state information of the control loop; and

[0141] A fault cause of the control loop is determined according to the status information.

[0142] The status information includes:

[0143] A linear relationship between the opening setting value SP of the control valve and the actual output value MV of the controlled variable associated with the control valve;

[0144] The deviation unilateral rate and output saturation rate related to the measured value PV, the set value SP and the output value MV of the control loop;

[0145] The similarity between the measured value PV of the control loop and the measured estimate; and / or

[0146] Deviation time series data of the control loop.

[0147] The linear relationship between the opening setting value SP of the control valve and the actual output value MV of the controlled variable related to the control valve can be calculated according to the following formula:

[0148]

[0149] Index=MATCH(Max(|ρ (n) |))

[0150] Among them, ρ (n) represents the sequence of correlation coefficients between the opening setting value of the control valve and the actual output value of the controlled variable associated with the control valve, PV (i) represents the measured value at the i-th moment, MV (i) Represents the valve position value at the i-th moment, Index is |ρ (n) |The location of the maximum value.

[0151] Among them, in the Index and ρ (Index) When the coefficient is within the preset range, it is determined that the control valve is in a normal working state; when the Index and ρ (Index) When the preset coefficient range is exceeded, it is determined that the control valve is faulty.

[0152] The deviation unilateral rate and output saturation rate related to the measured value PV, set value SP and output value MV of the control loop are calculated according to the following formula:

[0153]

[0154]

[0155] Deviation unilateral rate = max (deviation positive rate, deviation negative rate)

[0156]

[0157] Among them, e is the deviation value between the measured value PV and the set value SP of the auxiliary equipment. The deviation value e has a positive value or a negative value, which causes the output value MV of the auxiliary equipment to have a maximum value and a minimum value. When the deviation unilateral rate is greater than the set unilateral rate and the output saturation rate is greater than the set saturation rate, it is judged that the auxiliary equipment has a fault.

[0158] The auxiliary equipment includes one or more of the following: a heat exchanger, an air condenser and an electric heater.

[0159] Determining the similarity between the measured value PV of the control loop and the estimated value of the measured value PV according to a preset NARX neural network instrument valuation model;

[0160] The output value MV of the control loop is used as the NARX neural network instrument valuation model, the output value MV of the control loop and the predicted control loop correction factor α are used as input variables of the NARX neural network instrument valuation model, the measured value PV is used as the output variable to be estimated, and the output value MV and the correction factor α are input into the NARX neural network instrument valuation model to obtain the estimated value PVg of the measured value PV;

[0161] When the similarity between the estimated value PVg and the measured value PV is less than a set similarity, it is determined that the measuring instrument has a fault.

[0162] The similarity between the measured value PV and the estimated value PVg is calculated according to the following formula:

[0163]

[0164] Among them, PVi Represents the measured value at time i in the time series, PVg i It represents the predicted value at time i in the time series, and n represents the number of moments in the time series.

[0165] The deviation time series data of the control loop is calculated according to the following formula:

[0166] e i =PV i -SP i

[0167]

[0168]

[0169] Among them, PV i is the measured value at time i in the time series, SP i is the set value at time i in the time series, e i PV is the measured value at time i in the time series i With the set value SP i The deviation between i-j PV is the measured value PV of the time series at time i delayed by j time intervals i With the set value SP i The deviation between j is the autocorrelation coefficient of the deviation time series with itself and the time intervals delayed by j, and θ is the arithmetic mean of the autocorrelation coefficients delayed by 1 to m time intervals;

[0170] Among them, on the basis of eliminating control valve, equipment and instrument failures, when the θ value is close to 0, it is judged that the PID parameter model is well set; when the θ value is close to 1, it is judged that there is a problem with the PID parameter model setting, and parameter optimization or model adjustment is required.

[0171] An embodiment of the present invention provides a storage medium on which a program is stored. When the program is executed by a processor, the fault diagnosis method of the control loop of the refining device is implemented.

[0172] An embodiment of the present invention provides a processor, which is used to run a program, wherein the fault diagnosis method of the control loop of the refining device is executed when the program is running.

[0173] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0174] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0175] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0176] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0177] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0178] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0179] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0180] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0181] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for diagnosing a fault in a control loop of a refinery, characterized in that: The method comprises: Obtaining historical data parameters of one or more of the following in the control loop: a control valve, an auxiliary device, a measuring instrument, and a PID parameter model, wherein the historical data parameters include: a set value SP, a measured value PV, and an output value MV of the control loop; Calculating the state information of the control loop according to the acquired historical data parameters; and determining a fault cause of the control loop according to the state information, The status information includes: A linear relationship between the opening setting value SP of the control valve and the actual output value MV of the controlled variable associated with the control valve; The deviation unilateral rate and output saturation rate related to the measured value PV, the set value SP and the output value MV of the control loop; The similarity between the measured value PV of the control loop and the measured estimate; and / or Deviation time series data of the control loop.

2. The method according to claim 1, characterized in that The correlation coefficient between the opening setting value SP of the control valve and the actual output value MV of the controlled variable related to the control valve is calculated according to the following formula: Index=MATCH(Max(|ρ (n) |)) Among them, ρ (n) represents the sequence of correlation coefficients between the opening setting value of the control valve and the actual output value of the controlled variable associated with the control valve, PV (i) represents the measured value at the i-th moment, MV (i) Represents the valve position value at the i-th moment, Index is |ρ (n) |The location of the maximum value; Among them, in the Index and ρ (Index) When the coefficient is within the preset range, it is determined that the control valve is in a normal working state; when the Index and ρ (Index) When the preset coefficient range is exceeded, it is determined that the control valve is faulty.

3. The method according to claim 1, characterized in that The deviation unilateral rate and output saturation rate related to the measured value PV, set value SP and output value MV of the control loop are calculated according to the following formula: Deviation unilateral rate = max (deviation positive rate, deviation negative rate) Among them, e is the deviation value between the measured value PV and the set value SP of the auxiliary equipment. The deviation value e has a positive value or a negative value, which causes the output value MV of the auxiliary equipment to have a maximum value and a minimum value. When the deviation unilateral rate is greater than the set unilateral rate and the output saturation rate is greater than the set saturation rate, it is judged that the auxiliary equipment has a fault.

4. The method according to claim 1 or 3, characterized in that: The auxiliary equipment includes one or more of the following: a heat exchanger, an air condenser and an electric heater.

5. The method according to claim 1, characterized in that: Determining the similarity between the measured value PV of the control loop and the estimated value of the measured value PV according to a preset NARX neural network instrument valuation model; The output value MV of the control loop is used as the NARX neural network instrument valuation model, the output value MV of the control loop and the predicted control loop correction factor α are used as input variables of the NARX neural network instrument valuation model, the measured value PV is used as the output variable to be estimated, and the output value MV and the correction factor α are input into the NARX neural network instrument valuation model to obtain the estimated value PVg of the measured value PV; When the similarity between the estimated value PVg and the measured value PV is less than a set similarity, it is determined that the measuring instrument has a fault.

6. The method according to claim 5, characterized in that The similarity between the measured value PV and the estimated value PVg is calculated according to the following formula: Among them, PV i Represents the measured value at time i in the time series, PVg i It represents the predicted value at time i in the time series, and n represents the number of moments in the time series.

7. The method according to claim 1, characterized in that The deviation time series data of the control loop is calculated according to the following formula: and i =PV i -SP i Among them, PV i is the measured value at time i in the time series, SP i is the set value at time i in the time series, e i PV is the measured value at time i in the time series i With the set value SP i The deviation between i-j PV is the measured value PV of the time series at time i delayed by j time intervals i With the set value SP i The deviation between j is the autocorrelation coefficient of the deviation time series with itself and the time intervals delayed by j, and θ is the arithmetic mean of the autocorrelation coefficients delayed by 1 to m time intervals; Among them, on the basis of eliminating control valve, equipment and instrument failures, when the θ value is close to 0, it is judged that the PID parameter model is well set; when the θ value is close to 1, it is judged that there is a problem with the PID parameter model setting, and parameter optimization or model adjustment is required.

8. A fault diagnosis device for a control circuit of a refinery, characterized in that: The device comprises: A data acquisition unit, used to acquire historical data parameters of one or more of the following in the control loop: a control valve, an auxiliary device, a measuring instrument, and a PID parameter model, wherein the historical data parameters include: a set value SP, a measured value PV, and an output value MV of the control loop; A fault diagnosis unit is used to perform the following operations according to the historical data parameters acquired by the data acquisition unit: calculating state information of the control loop; and determining a fault cause of the control loop according to the state information, The status information includes: A linear relationship between the opening setting value SP of the control valve and the actual output value MV of the controlled variable associated with the control valve; The deviation unilateral rate and output saturation rate related to the measured value PV, the set value SP and the output value MV of the control loop; The similarity between the measured value PV of the control loop and the measured estimate; and / or Deviation time series data of the control loop.

9. The device according to claim 8, characterized in that The linear relationship between the opening setting value SP of the control valve and the actual output value MV of the controlled variable related to the control valve is calculated according to the following formula: Index=MATCH(Max(|ρ (n) |)) Among them, ρ (n) represents the sequence of correlation coefficients between the opening setting value of the control valve and the actual output value of the controlled variable associated with the control valve, PV (i) represents the measured value at the i-th moment, MV (i) Represents the valve position value at the i-th moment, Index is |ρ (n) |The location of the maximum value; Among them, in the Index and ρ (Index) When the coefficient is within the preset range, it is determined that the control valve is in a normal working state; when the Index and ρ (Index) When the preset coefficient range is exceeded, it is determined that the control valve is faulty.

10. The device according to claim 8, characterized in that The deviation unilateral rate and output saturation rate related to the measured value PV, set value SP and output value MV of the control loop are calculated according to the following formula: Deviation unilateral rate = max (deviation positive rate, deviation negative rate) Among them, e is the deviation value between the measured value PV and the set value SP of the auxiliary equipment. The deviation value e has a positive value or a negative value, which causes the output value MV of the auxiliary equipment to have a maximum value and a minimum value. When the deviation unilateral rate is greater than the set unilateral rate and the output saturation rate is greater than the set saturation rate, it is judged that the auxiliary equipment has a fault.

11. The device according to claim 8 or 10, characterized in that The auxiliary equipment includes one or more of the following: a heat exchanger, an air condenser and an electric heater.

12. The device according to claim 8, characterized in that Determining the similarity between the measured value PV of the control loop and the estimated value of the measured value PV according to a preset NARX neural network instrument valuation model; The output value MV of the control loop is used as the NARX neural network instrument valuation model, the output value MV of the control loop and the predicted control loop correction factor α are used as input variables of the NARX neural network instrument valuation model, the measured value PV is used as the output variable to be estimated, and the output value MV and the correction factor α are input into the NARX neural network instrument valuation model to obtain the estimated value PVg of the measured value PV; When the similarity between the estimated value PVg and the measured value PV is less than a set similarity, it is determined that the measuring instrument has a fault.

13. The device according to claim 12, characterized in that The similarity between the measured value PV and the estimated value PVg is calculated according to the following formula: Among them, PV i Represents the measured value at time i in the time series, PVg i It represents the predicted value at time i in the time series, and n represents the number of moments in the time series.

14. The device according to claim 8, characterized in that The deviation time series data of the control loop is calculated according to the following formula: and i =PV i -SP i Among them, PV i is the measured value at time i in the time series, SP i is the set value at time i in the time series, e i PV is the measured value at time i in the time series i With the set value SP i The deviation between i-j PV is the measured value PV of the time series at time i delayed by j time intervals i With the set value SP i The deviation between j is the autocorrelation coefficient of the deviation time series with itself and the time intervals delayed by j, and θ is the arithmetic mean of the autocorrelation coefficients delayed by 1 to m time intervals; Among them, on the basis of eliminating control valve, equipment and instrument failures, when the θ value is close to 0, it is judged that the PID parameter model is well set; when the θ value is close to 1, it is judged that there is a problem with the PID parameter model setting, and parameter optimization or model adjustment is required.

Images