Valve viscosity identification method and system

Abstract

The invention provides a valve viscosity identification method and system, and belongs to the technical field of automatic control. The method comprises the steps of obtaining a control deviation sequence and a valve output deviation sequence based on historical operation data of a target control loop; based on the control deviation sequence, a control deviation oscillation rate and a control deviation abnormal waveform oscillation rate are obtained through calculation; based on the valve output deviation sequence, a valve output deviation oscillation rate and a valve output deviation abnormal waveform oscillation rate are obtained through calculation; and on the basis of the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, valve viscosity risk judgment is conducted, and a risk judgment result is obtained. Therefore, support is provided for automatic valve viscosity diagnosis and valve maintenance, optimization and improvement of a control loop of the refining device are facilitated, stable operation of the refining device is guaranteed, and model training is not needed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic control, in particular to a valve stick identification method, a valve stick identification system, a machine readable storage medium and an electronic device. BACKGROUND

[0002] In the DCS control system of a refining and chemical plant, a large number of control loops are usually established to ensure stable operation of the entire system. These control loops can accurately control various process parameters such as temperature, pressure, liquid level, etc. These control loops play a crucial role in the refining and chemical plant, and through automatic control, production efficiency can be improved, energy consumption can be reduced, and human operation errors can be reduced, thereby ensuring production safety and product quality.

[0003] Valves, as common output units of control loops, will cause nonlinear faults due to the large static friction between the valve stem and the packing after a period of operation. In pneumatic control valves, this stickiness can cause limit cycles in the loop, causing the controlled quantity to oscillate up and down around the set value.

[0004] Valve stickiness in the control loop can easily weaken the control performance of the controller. Even without external disturbances, stable limit cycles can be generated, resulting in self-excited oscillation motion with fixed frequency and period. Oscillation is a serious manifestation of the deterioration of control loop performance, and it is relatively easy to spread to multiple control loops associated with it through coupling, and it may even cause plant-level oscillation. Because of the existence of stable limit cycles, the loop will continue to oscillate, so that the oscillation signal contains harmonic components, and the waveform shown by the oscillation signal may not be a simple sinusoidal wave form, but a periodic triangular wave or square wave form, which will cause greater fluctuations in the process variable, not only affecting the production of high-quality products in the process industry, but also increasing energy consumption, wasting raw materials, and accelerating equipment wear and tear, ultimately leading to a decline in product quality, yield, and economic efficiency.

[0005] The data-driven valve stickiness diagnosis is a method of diagnosing valve stickiness by using data analysis and machine learning technology. A classifier or regression model is constructed by using machine learning algorithms such as support vector machine, neural network, etc., to realize automatic diagnosis of valve stickiness. The model is evaluated using a test data set, and the accuracy, recall rate, etc. of the model are calculated to evaluate the performance of the model. The automatic diagnosis of valve stickiness is carried out by real-time monitoring of data, and the corresponding adjustment and maintenance are carried out according to the diagnosis result. The data-driven valve stickiness diagnosis method can automatically identify and predict valve stickiness failure by using a large amount of historical data and advanced machine learning technology, and improve the stability and reliability of the control loop. At the same time, this method can also realize the diagnosis of different types and models of valves, and has wide application prospect. However, most of these diagnosis methods need to establish a diagnosis model and need a large amount of data for training, and the abnormal data in the production process is generally less and presents different fault patterns.

[0006] Therefore, how to diagnose valve stickiness according to the periodic triangular wave or square wave abnormal oscillation in the automatic state of the control loop is a problem that needs to be solved urgently. SUMMARY

[0007] The purpose of the embodiments of the present application is to provide a valve stickiness identification method and system to at least solve the problem of not being able to diagnose valve stickiness according to the abnormal oscillation in the automatic state of the control loop.

[0008] To achieve the above-mentioned purpose, the first aspect of the present application provides a valve stickiness identification method, comprising:

[0009] Based on the historical operation data of the target control loop, a control deviation sequence and a valve output deviation sequence are obtained;

[0010] Based on the control deviation sequence, the control deviation oscillation rate and the control deviation abnormal waveform oscillation rate are calculated;

[0011] Based on the valve output deviation sequence, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate are calculated;

[0012] Based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, the risk of valve stickiness is judged, and the risk judgment result is obtained.

[0013] Optionally, the historical operation data of the target control loop includes a set value sequence SP[I] of the target control loop, a measured value sequence PV[I] and an output value sequence OP[I], I={1, 2,..., n}, n is the number of data collection;

[0014] The historical operation data of the target control loop is used to obtain a control deviation sequence and a valve output deviation sequence, including:

[0015] A control deviation sequence e[I] is obtained according to the deviation between a measured value sequence PV[I] and a set value sequence SP[I]; wherein e[I] = PV[I] - SP[I];

[0016] A valve output deviation sequence OPe[I] is obtained according to the deviation between an output value sequence OP[I] and the average output value of the output value sequence OP[I]; wherein,

[0017] Optionally, the control deviation oscillation rate and the control deviation abnormal waveform oscillation rate are calculated based on the control deviation sequence, including:

[0018] The control deviation sequence is classified and valued to obtain a plurality of first array data sequences;

[0019] Based on the minimum distance method, the average value and the average value similarity rate of each first array data sequence are calculated;

[0020] Based on the average value and the average value similarity rate of each first array data sequence, the abnormal waveform oscillation is identified, and the control deviation oscillation rate and the oscillation rate of each abnormal waveform are calculated; wherein the abnormal waveform includes a square waveform and a triangular waveform;

[0021] Based on the oscillation rate of each abnormal waveform, the control deviation abnormal waveform oscillation rate is obtained.

[0022] Optionally, the control deviation sequence is classified and valued to obtain a plurality of first array data sequences, including:

[0023] The control deviation sequence is traversed, and based on the data positive and negative change information of the control deviation sequence, each first array data sequence is filled in a pre-established array to obtain a plurality of first array data sequences;

[0024] The plurality of first array data sequences include a continuous positive deviation sequence, a continuous positive sum sequence, a continuous positive duration sequence, a continuous positive maximum value time point sequence, a continuous negative deviation sequence, a continuous negative sum sequence, a continuous negative duration sequence, and a continuous negative maximum value time point sequence.

[0025] Optionally, the control deviation sequence is classified and valued to obtain a plurality of first array data sequences, including:

[0026] In a case that the data of the first sub-sequence of the control deviation sequence is continuously positive and the data adjacent to the last data of the first sub-sequence is negative, the data of the first sub-sequence is stored to the continuous positive deviation sequence, the positive sum of the data of the first sub-sequence is stored to the continuous positive sum sequence, the acquisition time point corresponding to each data of the first sub-sequence is stored to the continuous positive duration sequence, and the acquisition time point satisfying the positive maximum value time point in the first sub-sequence is stored to the continuous positive maximum value time point sequence; wherein the positive maximum value time point represents all acquisition time points from the acquisition time point of the first data of the first sub-sequence to the acquisition time point corresponding to the maximum value of the data in the first sub-sequence.

[0027] In a case that the data of the first sub-sequence of the control deviation sequence is continuously negative and the data adjacent to the last data of the first sub-sequence is positive, the data of the first sub-sequence is stored to the continuous negative deviation sequence, the negative sum of the data of the first sub-sequence is stored to the continuous negative sum sequence, the acquisition time point corresponding to each data of the first sub-sequence is stored to the continuous negative duration sequence, and the acquisition time point satisfying the negative maximum value time point in the first sub-sequence is stored to the continuous negative maximum value time point sequence; wherein the negative maximum value time point represents all acquisition time points from the acquisition time point of the first data of the first sub-sequence to the acquisition time point corresponding to the negative maximum value of the data in the first sub-sequence; wherein the control deviation sequence is e[I], I={1, 2, …, n}, the first sub-sequence is e[i], i={a, a+1, …, b}, a≥1, and b is less than n.

[0028] Optionally, the minimum distance method is used to calculate the average value and the average value similarity rate of each first array data sequence, including:

[0029] According to the formula the average value K[imin] corresponding to each first array data sequence is determined; wherein K[i] represents the data corresponding to the i moment in the first array data sequence, K[j] represents the data corresponding to the j moment in the first array data sequence, K[imin] represents the data corresponding to the imin moment in the first array data sequence, and imin is the moment corresponding to the minimum value of the square sum of the deviation between all data of the first array data sequence and K[imin] is not more than the square sum of the deviation between all data of the first array data sequence and K[j]; wherein j=1, 2, …, m, i=1, 2, …, m, and m is greater than 10.

[0030] ​For each first array data sequence, judging and removing data in the first array data sequence that is not similar to the corresponding average value K[imin], to obtain a first new array data sequence Knew[I] that is similar to the average value K[imin]; wherein, I={1, 2, …, z}, z≤m;

[0031] Based on each first new array data sequence, calculating a corresponding latest average value Avgnew and average value similarity rate.

[0032] Optionally, the judging and removing data in the first array data sequence that is not similar to the corresponding average value K[imin], to obtain the first new array data sequence Knew[I] that is similar to the average value K[imin] includes:

[0033] When the average value K[imin] is zero, the parameter z is zero, so that the corresponding first new array data sequence Knew[I] is an empty sequence;

[0034] When or , it indicates that the value K[i] is not similar to the corresponding average value K[imin], and the data K[i] in the first array data sequence at the i moment is removed, to obtain the first new array data sequence Knew[I] that is similar to the average value K[imin]; wherein,

[0035] When the parameter z is less than a preset value, it is determined that the corresponding average value similarity rate is zero, i=1, 2, …, m.

[0036] Optionally, the calculation formula of the average value similarity rate is as follows:

[0037]

[0038] Wherein, si represents the average value similarity rate, and Avgnew represents the latest average value.

[0039] Optionally, the above-mentioned average value and average value similarity rate based on each first array data sequence are used to identify abnormal waveform oscillation, including:

[0040] Combining the average value and the average value similarity rate of each first array data sequence to obtain a corresponding combination decision set;

[0041] Based on a preset waveform oscillation determination rule library, matching a waveform oscillation determination result corresponding to the combination decision set; wherein,

[0042] The preset waveform oscillation determination rule library stores determination rules corresponding to all waveform oscillation determination results.

[0043] Optionally, the first array data sequence includes a continuous positive deviation sequence and a continuous negative deviation sequence;

[0044] The calculation formula of the oscillation rate of the square waveform is as follows:

[0045]

[0046] Wherein, FRateF represents the oscillation rate of the square waveform, fASOsc represents the average value similarity rate of the continuous positive deviation sequence, and fBSOsc represents the average value similarity rate of the continuous negative deviation sequence.

[0047] Optionally, the first array data sequence includes a continuous positive maximum value time point sequence and a continuous negative maximum value time point sequence.

[0048] The calculation formula of the oscillation rate of the triangular waveform is as follows:

[0049]

[0050] Wherein, FRateS represents the oscillation rate of the triangular waveform, AMTavg represents the average value of the continuous positive maximum value time point sequence, and BMTavg represents the average value of the continuous negative maximum value time point sequence.

[0051] Optionally, the oscillation rate of the control deviation abnormal waveform is obtained based on the oscillation rate of each abnormal waveform, and the method comprises the following steps:

[0052] Taking the maximum value in the oscillation rates of all abnormal waveforms as the oscillation rate of the control deviation abnormal waveform; wherein,

[0053] eFRate = max (FRateF, FRateS), eFRate represents the oscillation rate of the control deviation abnormal waveform, FRateF represents the oscillation rate of the square waveform, and FRateS represents the oscillation rate of the triangular waveform.

[0054] Optionally, the first array data sequence includes a continuous positive sum sequence, a continuous positive duration sequence, a continuous negative sum sequence, and a continuous negative duration sequence.

[0055] The calculation formula of the control deviation oscillation rate is as follows:

[0056]

[0057] Wherein, eRate represents the control deviation oscillation rate, ASOsc represents the average value similarity rate of the continuous positive sum sequence, ATOsc represents the average value similarity rate of the continuous positive duration sequence, BSOsc represents the average value similarity rate of the continuous negative sum sequence, and BTOsc represents the average value similarity rate of the continuous negative duration sequence.

[0058] Optionally, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate are calculated based on the valve output deviation sequence, and the calculation comprises:

[0059] The valve output deviation sequence is classified and valued to obtain a plurality of second array data sequences.

[0060] Based on the minimum distance method, the average value and the average value similarity rate of each second array data sequence are calculated.

[0061] Based on the average value and the average value similarity rate of each second array data sequence, the abnormal waveform oscillation is identified, and the valve output deviation oscillation rate and the oscillation rate of each abnormal waveform are calculated; wherein the abnormal waveform includes a square waveform and a triangular waveform.

[0062] Based on the oscillation rate of each abnormal waveform, the valve output deviation abnormal waveform oscillation rate is obtained.

[0063] Optionally, the valve sticking risk is judged based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, and the judgment comprises:

[0064] Based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, the valve sticking rate St is obtained.

[0065] Based on the valve sticking rate St, the valve sticking risk is judged; wherein the closer the valve sticking rate St is to 100%, the higher the positive degree of the valve sticking is, and when the valve sticking rate St is greater than 80%, a valve sticking risk signal is generated.

[0066] Optionally, the calculation formula of the valve sticking rate St is as follows:

[0067]

[0068] Wherein, eRate represents the control deviation oscillation rate, eFRate represents the control deviation abnormal waveform oscillation rate, OPRate represents the valve output deviation oscillation rate, and OPFRate represents the valve output deviation abnormal waveform oscillation rate.

[0069] Optionally, the valve sticking identification method further comprises:

[0070] The historical operation data of the target control loop is obtained through an industrial numerical sampling interface and / or a database interface.

[0071] The second aspect of the application provides a valve sticking identification system, comprising:

[0072] The deviation sequence obtaining module is configured to obtain a control deviation sequence and a valve output deviation sequence based on historical operation data of the target control loop.

[0073] The control deviation oscillation rate calculation module is configured to calculate a control deviation oscillation rate and a control deviation abnormal waveform oscillation rate based on the control deviation sequence.

[0074] The valve output deviation oscillation rate calculation module is configured to calculate a valve output deviation oscillation rate and a valve output deviation abnormal waveform oscillation rate based on the valve output deviation sequence.

[0075] The valve stickiness risk judgment module is configured to perform a risk judgment of valve stickiness based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate, and the valve output deviation abnormal waveform oscillation rate, and obtain a risk judgment result.

[0076] In a third aspect of the present application, a machine-readable storage medium is provided, which stores instructions configured to cause a processor to be configured to perform the valve stickiness identification method described above when the instructions are executed by the processor.

[0077] In a fourth aspect of the present application, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the valve stickiness identification method described above when executing the computer program.

[0078] By the above technical solution, the valve stickiness identification method and system are provided, which obtain a control deviation sequence and a valve output deviation sequence based on historical operation data of a target control loop, analyze and calculate oscillation rates and abnormal oscillation conditions of the control deviation sequence and the valve output deviation sequence respectively, obtain a control deviation oscillation rate, a control deviation abnormal waveform oscillation rate, a valve output deviation oscillation rate, and a valve output deviation abnormal waveform oscillation rate, and then quickly determine a valve stickiness condition based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate, and the valve output deviation abnormal waveform oscillation rate, thereby providing support for automatic diagnosis of valve stickiness and maintenance of valves, and being conducive to optimization and improvement of a control loop of a refining device, and ensuring smooth operation of the refining device. The method and system use historical operation data of measured values, set values, and output values of a real control loop of the refining device to identify abnormal oscillation waveforms, thereby quickly identifying valve stickiness without model training.

[0079] Other features and advantages of the embodiments of the present application will be described in detail in the following detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0080] The accompanying drawings are included to provide a further understanding of embodiments of the application, and are incorporated in and constitute a part of this specification, illustrate embodiments of the application, and together with the description serve to explain embodiments of the application, but are not intended to limit the present application in any manner. In the drawings:

[0081] Figure 1 is a flow chart of a valve stickiness identification method provided by an embodiment of the present application;

[0082] Figure 2 is a block diagram of a valve stickiness identification system provided by an embodiment of the present application;

[0083] Figure 3 is a valve stickiness curve diagram provided by an embodiment of the present application;

[0084] Figure 4 is a flow chart of another valve stickiness identification method provided by an embodiment of the present application;

[0085] Figure 5 is a schematic structural diagram of an electronic device provided by a preferred embodiment of the present application.

[0086] Legend of reference signs

[0087] 10 - electronic device, 100 - processor, 101 - memory, 102 - computer program. DETAILED DESCRIPTION

[0088] The specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely intended to illustrate and explain the present application, and are not intended to limit the present application.

[0089] Example 1

[0090] Figure 1 is a flow chart of a valve stickiness identification method provided by an embodiment of the present application. As shown in Figure 1 , the embodiment of the present application provides a valve stickiness identification method, comprising:

[0091] S110: obtaining a control deviation sequence and a valve output deviation sequence based on historical operation data of a target control loop;

[0092] In some embodiments of the present embodiment, the historical operation data of the target control loop includes a set value sequence SP[I], a measured value sequence PV[I] and an output value sequence OP[I] of the target control loop, I={1, 2,..., n}, n is the number of data collection; the control deviation sequence and the valve output deviation sequence are obtained based on the historical operation data of the target control loop, including: obtaining the control deviation sequence e[I] according to the deviation between the measured value sequence PV[I] and the set value sequence SP[I]; wherein e[I]=PV[I]-SP[I]; obtaining the valve output deviation sequence OPe[I] according to the deviation between the output value sequence OP[I] and the average output value of the output value sequence OP[I]; wherein,

[0093] S120: based on the control deviation sequence, the control deviation oscillation rate and the control deviation abnormal waveform oscillation rate are calculated;

[0094] In some embodiments of the present embodiment, the control deviation oscillation rate and the control deviation abnormal waveform oscillation rate are calculated based on the control deviation sequence, including:

[0095] The control deviation sequence is classified and valued to obtain a plurality of first array data sequences;

[0096] In detail, the control deviation sequence is classified and valued to obtain a plurality of first array data sequences, including: traversing the control deviation sequence, based on the data positive and negative change information of the control deviation sequence, filling the pre-established array corresponding to each first array data sequence to obtain a plurality of first array data sequences; wherein the plurality of first array data sequences include a continuous positive deviation sequence, a continuous positive and sequence, a continuous positive duration sequence, a continuous positive maximum value time point sequence, a continuous negative deviation sequence, a continuous negative and sequence, a continuous negative duration sequence and a continuous negative maximum value time point sequence.

[0097] Further, the above data positive and negative change information based on the control deviation sequence is used to fill the pre-established array corresponding to each first array data sequence to obtain a plurality of first array data sequences, including: in the case that the data of the first sub-sequence of the control deviation sequence is continuously positive and the data adjacent to the end data of the first sub-sequence is negative, storing the data of the first sub-sequence to the continuous positive deviation sequence, storing the positive sum of the data of the first sub-sequence to the continuous positive sum sequence, storing the collection time point corresponding to each data of the first sub-sequence to the continuous positive duration sequence, and storing the collection time point satisfying the positive maximum value time point in the first sub-sequence to the continuous positive maximum value time point sequence; wherein the positive maximum value time point represents all collection time points from the collection time point of the first data of the first sub-sequence to the collection time point corresponding to the maximum value of the data in the first sub-sequence; in the case that the data of the first sub-sequence of the control deviation sequence is continuously negative and the data adjacent to the end data of the first sub-sequence is positive, storing the data of the first sub-sequence to the continuous negative deviation sequence, storing the negative sum of the data of the first sub-sequence to the continuous negative sum sequence, storing the collection time point corresponding to each data of the first sub-sequence to the continuous negative duration sequence, and storing the collection time point satisfying the negative maximum value time point in the first sub-sequence to the continuous negative maximum value time point sequence; wherein the negative maximum value time point represents all collection time points from the collection time point of the first data of the first sub-sequence to the collection time point corresponding to the negative maximum value of the data in the first sub-sequence; wherein the control deviation sequence is e[I], I={1, 2,..., n}, and the first sub-sequence is e[i], i={a, a+1,..., b}, a≥1, and b is less than n.

[0098] Specifically, the control deviation sequence e[I] (wherein I={1, 2,..., n}) is traversed from 1 to n, the pre-established array is filled according to the data positive and negative change information of the control deviation sequence according to method (1) and method (2) to obtain a plurality of first array data sequences, and the method (1) and the method (2) are as follows:

[0099] Method (1): if the data of e[I] (I=a, a+1,..., b) is continuously positive and e[b+1] is negative, e[c]∈e[I] (I=a, a+1,..., b), e[c] represents the maximum value in e[I] (I=a, a+1,..., b), and c is the time point corresponding to the maximum value of the data sequence e[I] (I=a, a+1,..., b), then: storing the positive e[I] (I=a, a+1,..., b) to the continuous positive deviation sequence, storing the continuous positive sum to the continuous positive sum sequence, and storing the collection time point corresponding to each data of the first sub-sequence to the continuous positive duration sequence. The continuous positive duration sequence is used to store the continuous positive duration ba (ie, all time points from b to a), and the continuous positive maximum time point sequence is used to store the continuous positive maximum time point ca (ie, all time points from c to a).

[0100] Method (2): If the data of e[I](I=a, a+1, ..., b) are continuously negative and e[b+1] is positive, e[c]∈e[I](I=a, a+1, ..., b), e[c] represents the minimum value in e[I](I=a, a+1, ..., b), and c is the time corresponding to the minimum value of the data sequence e[I](I=a, a+1, ..., b), then: use the continuous negative deviation sequence to store the negative |e[I]|(I=a, a+1, ..., b); use the continuous negative sum sequence to store the continuous negative sum The continuous negative duration sequence is used to store the continuous negative duration ba (ie, all time points from b to a); the continuous negative maximum time point sequence is used to store the continuous negative maximum time point ca (ie, all time points from c to a).

[0101] Based on the minimum distance method, the average value and the average value similarity are calculated for each first array data sequence respectively;

[0102] In detail, the above method based on the minimum distance is used to calculate the average value and the average similarity of each first array data sequence, including: according to the formula Determine the average value K[imin] corresponding to each first array data sequence; where K[i] represents the data corresponding to time i in the first array data sequence, K[j] represents the data corresponding to time j in the first array data sequence, K[imin] represents the data corresponding to time imin in the first array data sequence, and imin is The moment corresponding to the minimum value of Indicates that the sum of the squares of deviations between all data of the first array data sequence and K[imin] does not exceed the sum of the squares of deviations between all data of the first array data sequence and K[j]; wherein j = 1, 2, ..., m, i = 1, 2, ..., m, and m is greater than 10; for each first array data sequence, determine and clear the data in the first array data sequence that are not close to the corresponding average value K[imin], and obtain a first new array data sequence Knew[I] that is close to the average value K[imin]; wherein I = {1, 2, ..., z}, z≤m; based on each first new array data sequence, calculate the corresponding latest average value Avgnew and the average value similarity.

[0103] Further, the above judging and removing the data in the first array data sequence which is not similar to the average value K[imin] obtains the first new array data sequence Knew[I] which is similar to the average value K[imin], including: when the average value K[imin] is zero, setting the parameter z as zero so as to make the corresponding first new array data sequence Knew[I] as an empty sequence; when or , it indicates that the value of K[i] is not similar to the corresponding average value K[imin], and the data K[i] in the first array data sequence at the i moment is removed to obtain the first new array data sequence Knew[I] which is similar to the average value K[imin]; wherein, when the parameter z is less than a preset value, it is determined that the corresponding average value similarity rate is zero, i=1, 2,..., m.

[0104] In some embodiments of the embodiment, the calculation formula of the above average value similarity rate is as follows: Wherein, si indicates the average value similarity rate, and Avgnew indicates the latest average value.

[0105] Specifically, the process of calculating the average value and the average value similarity rate by using the minimum distance method is as follows: first, the first array data sequence K[I] which needs to calculate the average value is determined (wherein, I={1, 2,..., m}, and m is greater than 10), when at the imin moment, the sum of squares of deviations of all data of the first array data sequence K[I] from K[imin] is minimum, then K[imin] is the average value, that is The average value avg=K[imin] is obtained. If the average value MinAbs(K) is 0, it indicates that there is no fluctuation, z=0 is set to empty the corresponding first array data sequence K[I], and then the first new array data sequence Knew[I] obtained is an empty sequence. If or , it indicates that the value of K[i] is not similar to the corresponding average value K[imin], and the data K[i] in the first array data sequence at the i moment is removed to obtain the first new array data sequence Knew[I] which is similar to the average value K[imin]; wherein, when the parameter z is less than a preset value, it is determined that the corresponding average value similarity rate is zero, i=1, 2,..., m. The latest average value Avgnew is re-determined as avgnew = MinAbs(Knew). According to all data Knew[i] of the first new array data sequence Knew[I] and the latest average value Avgnew, the corresponding average value similarity rate is obtained based on the formula The corresponding average value similarity rate is obtained.

[0106] Based on the average value and the average value similarity rate of each first array data sequence, abnormal waveform oscillation is identified, and a control deviation oscillation rate and an oscillation rate of each abnormal waveform are calculated; wherein the abnormal waveform includes a square waveform and a triangular waveform.

[0107] In detail, the identification of the abnormal waveform oscillation based on the average value and the average value similarity rate of each first array data sequence includes: combining the average value and the average value similarity rate of each first array data sequence to obtain a corresponding combination decision set; and matching a waveform oscillation decision result corresponding to the combination decision set based on a preset waveform oscillation decision rule library; wherein the preset waveform oscillation decision rule library stores decision rules corresponding to all waveform oscillation decision results.

[0108] Specifically, if the average value similarity rate of the continuous positive deviation sum is too large (i.e. the average value similarity rate of the continuous positive sum sequence > 40) and the average value similarity rate of the continuous negative deviation sum is too large (the average value similarity rate of the continuous negative sum sequence > 40), it indicates that there may be oscillation, otherwise the oscillation rate Rate = 0 and the abnormal oscillation rate FRate = 0, and the diagnosis is exited. If the average value similarity rate of the continuous positive deviation is too large (the average value similarity rate of the continuous positive deviation sequence > 40) and the average value similarity rate of the continuous negative deviation is too large (the average value similarity rate of the continuous negative deviation sequence > 40), it indicates that there may be square waveform oscillation. If the average value similarity rate of the continuous positive maximum value duration is too large (the average value similarity rate of the continuous positive maximum value time point sequence > 40) and the average value similarity rate of the continuous negative minimum value duration is too large (the average value similarity rate of the continuous negative maximum value time point sequence > 40), it indicates that there is oscillation in time; if the deviation between the average value of the continuous positive maximum value time point sequence and the average value of the continuous negative maximum value time point sequence is too large, it indicates that there is abnormal fluctuation oscillation, such as triangular waveform oscillation.

[0109] In the above implementation process, the oscillation of the control deviation sequence is evaluated according to the similarity of the cumulative sum (i.e. the area) of each continuous positive or negative value corresponding to the control deviation sequence and the duration of each continuous positive or negative value. According to whether the area of the continuous positive or negative value is smooth, square oscillation waveform is diagnosed; according to the duration of the continuous positive value and the offset of the duration of the continuous positive value, triangular and other oscillation waveforms are diagnosed.

[0110] In some embodiments of the present embodiment, the first array data sequence comprises a continuous positive deviation sequence and a continuous negative deviation sequence; and the calculation formula of the oscillation rate of the square waveform is as follows: Wherein, FRateF represents the oscillation rate of the square waveform, fASOsc represents the average value similarity rate of the continuous positive deviation sequence, and fBSOsc represents the average value similarity rate of the continuous negative deviation sequence.

[0111] In some embodiments of the present embodiment, the first array data sequence comprises a continuous positive maximum value time point sequence and a continuous negative maximum value time point sequence; and the calculation formula of the oscillation rate of the triangular waveform is as follows: Wherein, FRateS represents the oscillation rate of the triangular waveform, AMTavg represents the average value of the continuous positive maximum value time point sequence, and BMTavg represents the average value of the continuous negative maximum value time point sequence.

[0112] In some embodiments of the present embodiment, the first array data sequence comprises a continuous positive sum sequence, a continuous positive duration sequence, a continuous negative sum sequence, and a continuous negative duration sequence; and the calculation formula of the control deviation oscillation rate is as follows: Wherein, eRate represents the control deviation oscillation rate, ASOsc represents the average value similarity rate of the continuous positive sum sequence, ATOsc represents the average value similarity rate of the continuous positive duration sequence, BSOsc represents the average value similarity rate of the continuous negative sum sequence, and BTOsc represents the average value similarity rate of the continuous negative duration sequence.

[0113] Specifically, the average value of the average value similarity rate of the continuous positive sum sequence, the average value similarity rate of the continuous positive duration sequence, the average value similarity rate of the continuous negative sum sequence, and the average value similarity rate of the continuous negative duration sequence is taken to obtain the control deviation oscillation rate.

[0114] Based on the oscillation rate of each abnormal waveform, the control deviation abnormal waveform oscillation rate is obtained.

[0115] In detail, based on the oscillation rate of each abnormal waveform, the control deviation abnormal waveform oscillation rate is obtained, including: taking the maximum value in the oscillation rate of all abnormal waveforms as the control deviation abnormal waveform oscillation rate; wherein, eFRate = max(FRateF, FRateS), eFRate represents the control deviation abnormal waveform oscillation rate, FRateF represents the oscillation rate of the square waveform, and FRateS represents the oscillation rate of the triangular waveform.

[0116] S130: based on the valve output deviation sequence, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate are calculated;

[0117] In detail, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate are calculated based on the valve output deviation sequence, including: classifying and assigning values to the valve output deviation sequence to obtain a plurality of second array data sequences; calculating the average value and the average value similarity rate of each second array data sequence based on the minimum distance method; identifying abnormal waveform oscillation based on the average value and the average value similarity rate of each second array data sequence, and calculating the valve output deviation oscillation rate and the oscillation rate of each abnormal waveform; wherein the abnormal waveform includes a square waveform and a triangular waveform; and obtaining the valve output deviation abnormal waveform oscillation rate based on the oscillation rate of each abnormal waveform.

[0118] It should be noted that the process of "calculating the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate" is consistent with the calculation process of "calculating the control deviation oscillation rate and the control deviation abnormal waveform oscillation rate", which will not be described here.

[0119] Specifically, the control deviation sequence and the valve output deviation sequence are calculated by S120 and S130, and the control deviation oscillation rate eRate, the control deviation abnormal waveform oscillation rate eFRate, the valve output deviation oscillation rate OPRate and the valve output deviation abnormal waveform oscillation rate OPFRate are obtained.

[0120] In the above implementation process, the method calculates the oscillation rate of the control loop control deviation and the valve opening operation data by area and duration, performs valve stick diagnosis based on abnormal waveforms such as square waveforms and triangular waveforms, and provides a basis for valve fault automatic diagnosis and valve maintenance.

[0121] S140: Based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, the risk of valve stick is judged, and the risk judgment result is obtained.

[0122] In detail, the risk of valve stick is judged based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, including: obtaining the valve stick rate St based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate; judging the risk of valve stick based on the valve stick rate St; wherein the closer the valve stick rate St is to 100%, the higher the degree of valve stick risk, and when the valve stick rate St is greater than 80%, it is considered that the valve has a stick risk, and a valve stick risk signal is generated to remind the user to check and maintain the valve in time.

[0123] In some embodiments of the present embodiment, the calculation formula of the valve stickiness St is as follows: Wherein, eRate represents the control deviation oscillation rate, eFRate represents the control deviation abnormal waveform oscillation rate, OPRate represents the valve output deviation oscillation rate, and OPFRate represents the valve output deviation abnormal waveform oscillation rate.

[0124] Specifically, the method obtains a control deviation sequence and a valve output deviation sequence based on historical operation data of a target control loop, analyzes and calculates the oscillation rate and abnormal oscillation of the control deviation sequence and the valve output deviation sequence respectively, obtains the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, and thus quickly determines the valve stickiness based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, provides support for automatic diagnosis of valve stickiness and maintenance of the valve, and is conducive to optimization and improvement of the control loop of the refining device, and ensures stable operation of the refining device. The method uses the historical operation data of the measured value, the set value and the output value of the real control loop of the refining device to identify the abnormal oscillation waveform, so as to quickly identify the valve stickiness without model training.

[0125] In some embodiments of the present embodiment, the valve stickiness identification method further comprises: obtaining the historical operation data of the target control loop through an industrial control data acquisition interface and / or a database interface.

[0126] Embodiment 2

[0127] Figure 2 is a block diagram of a valve stickiness identification system provided by an embodiment of the present application. As shown in Figure 2 The present embodiment provides a valve stickiness identification system, which comprises:

[0128] a deviation sequence obtaining module, configured to obtain a control deviation sequence and a valve output deviation sequence based on historical operation data of a target control loop;

[0129] a control deviation oscillation rate calculation module, configured to calculate a control deviation oscillation rate and a control deviation abnormal waveform oscillation rate based on the control deviation sequence;

[0130] a valve output deviation oscillation rate calculation module, configured to calculate a valve output deviation oscillation rate and a valve output deviation abnormal waveform oscillation rate based on the valve output deviation sequence;

[0131] a valve stickiness risk judgment module, configured to perform risk judgment of valve stickiness based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, and obtain a risk judgment result.

[0132] Specifically, the system obtains a control deviation sequence and a valve output deviation sequence based on historical operation data of a target control loop, analyzes and calculates oscillation rates and abnormal oscillation conditions of the control deviation sequence and the valve output deviation sequence respectively to obtain a control deviation oscillation rate, a control deviation abnormal waveform oscillation rate, a valve output deviation oscillation rate and a valve output deviation abnormal waveform oscillation rate, so as to quickly determine the valve sticking condition based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, provide support for automatic diagnosis of valve sticking and valve maintenance, and be beneficial to optimization and improvement of the control loop of the refining device, and ensure stable operation of the refining device. The system uses historical operation data of measured values, set values and output values of the real control loop of the refining device to identify abnormal oscillation waveforms, so as to quickly identify valve sticking without model training.

[0133] Embodiment 3

[0134] Figure 1 is a flowchart of a valve sticking identification method provided by an embodiment of the present application. As shown in Figure 1 , the embodiment of the present application provides a valve sticking identification method.

[0135] In this embodiment, before the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate are calculated, the control deviation sequence and the valve output deviation sequence need to be judged for abnormal data respectively. If there is no data fluctuation in the control deviation sequence or the valve output deviation sequence, the calculation of the oscillation rate and the abnormal waveform oscillation rate of the control deviation sequence or the valve output deviation sequence is exited. If the data amount in the control deviation sequence or the valve output deviation sequence is insufficient (i.e. the data amount does not meet the preset data amount requirement), the calculation of the oscillation rate and the abnormal waveform oscillation rate of the control deviation sequence or the valve output deviation sequence is also exited.

[0136] Embodiment 4

[0137] Figure 1 is a flowchart of a valve sticking identification method provided by an embodiment of the present application. As shown in Figure 1 , the embodiment of the present application provides a valve sticking identification method.

[0138] In this embodiment, the method can also obtain an oscillation period Period according to the average value ATavg of the average value of the duration of the continuous positive direction and the average value BTavg of the average value of the duration of the continuous negative direction, and the calculation formula is as follows:

[0139] Embodiment 5

[0140] Figure 3 is a valve stick curve provided by an embodiment of the present application. As shown in the figure, the oscillation waveform of the measured value is square wave oscillation, and the oscillation waveform of the output value is an abnormal triangle. Due to valve stick, the valve output value continuously increases (from 78.31 to 81.83), while the controlled parameter does not change (fluctuates around 83), the control loop continuously increases the valve output value, and when the critical value is reached (81.83), the valve opens and is too large, resulting in a sudden drop in the controlled variable (suddenly drops from around 83 to around 78.2), and due to the excessive adjustment, the control loop continuously reduces the valve output value, resulting in the abnormal waveform oscillation waveform. Figure 3

[0141] is a flow chart of another valve stick identification method provided by an embodiment of the present application. As shown in the figure, the embodiment provides a valve stick identification method, which comprises: obtaining control loop historical data; processing control deviation data and valve data; calculating oscillation rate and abnormal oscillation of control deviation and valve data respectively according to the oscillation monitoring method based on area and duration; and determining valve stick according to abnormal oscillation of control deviation and valve. Figure 4 Figure 4 The specific steps of the method are as follows:

[0142] Step 1: Obtain historical operation data of the control loop;

[0143] The historical operation data of the control loop can be obtained through an industrial control data acquisition interface or a database interface, including set value SP, measured value PV, output value OP, etc.

[0144] Step 2: Data processing of the control loop data;

[0145] 1) Obtain control deviation sequence e[i] (i=1, 2,..., n) according to the deviation of the measured value and the set value:

[0146] e[i]=PV[i]-SP[i];

[0147] 2) Obtain valve output deviation sequence OPe[i] (i=1, 2,..., n) according to the deviation of the output value sequence and the average value:

[0148]

[0149]

[0150] Step 3: Oscillation rate and abnormal waveform oscillation rate calculation of data sequence;

[0151] ​​The method uses the cumulative sum (area) of each continuous positive or negative value of the data sequence to evaluate the oscillation condition of the data sequence. According to whether the continuous positive or negative value is stable, the square oscillation waveform is diagnosed; according to the duration of the continuous positive value and the offset condition of the duration of the continuous positive value, the triangular oscillation waveform is diagnosed.

[0152] 1) The data sequence is classified and valued:

[0153] From 1 to n, the deviation sequence e[i] (i=1, 2,..., n) is traversed according to the positive and negative change of the data, and the related sequence is filled according to the methods (1) and (2):

[0154] (1) If e[i] (i=a, a+1,..., b) is continuously positive, e[b+1] is negative, and e[c]≥e[i] (i=a, a+1,..., b), then:

[0155] An array is established to store the positive e[i] (i=a, a+1,..., b);

[0156] An array is established to store the continuous positive sum

[0157] An array is established to store the continuous positive duration b-a;

[0158] An array is established to store the continuous positive maximum value time point c-a;

[0159] (2) If e[i] (i=a, a+1,..., b) is continuously negative, e[b+1] is positive, and e[c]≤e[i] (i=a, a+1,..., b), then:

[0160] An array is established to store the negative |e[i]| (i=a, a+1,..., b);

[0161] An array is established to store the continuous negative sum

[0162] An array is established to store the continuous negative duration b-a;

[0163] An array is established to store the continuous negative maximum value time point c-a;

[0164] 2) The average value and average value similarity rate of each data sequence are calculated by the minimum distance method:

[0165] (1) The minimum distance method calculates the average value MinAbs(K): the data sequence K[i] (i=1, 2,..., m) needs to be calculated, and m>10.

[0166] When at the moment imin, all data of K[i] (i=1, 2,..., m) has the minimum square sum of deviation from K[imin], K[imin] is the average value;

[0167]

[0168] The average value avg=K[imin];

[0169] (2) The minimum distance method is used to calculate the average value and the average value similarity rate: the data sequence K[i] (i=1, 2,..., m) needs to be calculated, and m>10.

[0170] (a) Obtain the average value avg=MinAbs(K);

[0171] (b) Remove the data in the data sequence K that is not similar to the average value;

[0172] If the average value is 0, it means no fluctuation, let z=0, and clear the K series data;

[0173] If the average value is not similar, remove the data of K series at the moment i;

[0174] Then the new data sequence Knew[i] (i=1, 2,..., z) similar to the average value is obtained.

[0175] If z is too small, exit, that is, the average value similarity rate is 0.

[0176] (c) Recalculate the average value Avgnew and the similarity rate si:

[0177] (3) The minimum distance method is used to calculate the average value and the average value similarity rate of each sequence respectively.

[0178] 3) Calculate the overall oscillation rate:

[0179] (1) If the average value similarity rate of the average value of the positive deviation sum is too large and the average value similarity rate of the average value of the negative deviation sum is too large, it means that there may be oscillation, otherwise exit the diagnosis;

[0180] (2) If the average value similarity rate of the average value of the positive deviation is too large and the average value similarity rate of the average value of the negative deviation is too large, it means that there may be square wave oscillation;

[0181] (3) If the average value similarity rate of the average value of the maximum value duration of the positive direction is too large and the average value similarity rate of the average value of the minimum value duration of the negative direction is too large, it means that there is oscillation in time; if the deviation between AMTavg and BMTavg is too large, it means that there is abnormal fluctuation oscillation, such as triangular wave oscillation;

[0182] (4) the maximum value of the square wave abnormal oscillation rate and the triangle wave abnormal oscillation rate is taken as the abnormal oscillation rate;

[0183] (5) the average value of the average value similarity rate ASOsc of the continuously positive deviation sum, the average value similarity rate ATOsc of the continuously positive duration, the average value similarity rate BSOsc of the continuously negative deviation sum, and the average value similarity rate BTOsc of the continuously negative duration is taken as the oscillation rate Rate;

[0184] (6) the average value of the average value ATavg of the continuously positive duration and the average value BTavg of the continuously negative duration is taken as the oscillation period Period.

[0185] Step 4: the oscillation rate and the abnormal wave oscillation rate of the control deviation and the valve output deviation are calculated;

[0186] The oscillation rate and the abnormal wave oscillation rate of the control deviation and the valve output deviation are calculated through step 3.

[0187] Step 5: the valve stickiness rate St is calculated;

[0188] The valve stickiness rate St is obtained through the oscillation rate and the abnormal wave oscillation rate of the control deviation and the valve output deviation.

[0189] Step 6: the valve stickiness rate is judged;

[0190] The closer the stickiness rate St is to 100%, the higher the valve stickiness degree is, and thus when the stickiness rate St is greater than 80%, it is considered that the valve has a stickiness risk and needs to be checked and maintained in time.

[0191] Through steps 1 to 6, the following can be calculated:

[0192] The oscillation rate of the control deviation is 95.14%, and the abnormal fluctuation oscillation rate is 98.56%;

[0193] The oscillation rate of the valve output deviation is 97.43%, and the abnormal fluctuation oscillation rate is 99.16%.

[0194] The valve stickiness rate is 94.09%, which is greater than 80%, and thus it is considered that the valve has stickiness.

[0195] Example 6

[0196] The embodiment of the application provides a machine readable storage medium, which stores instructions, and the instructions make the processor 100 be configured to execute the valve stickiness identification method when executed by the processor 100.

[0197] Machine-readable storage media includes 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 disc read-only memory (CD-ROM), digital versatile disc (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 transitory media such as modulated data signals and carrier waves.

[0198] An embodiment of the present invention further provides an electronic device 10, which includes a memory 101, a processor 100, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, the above-mentioned valve viscosity identification method is implemented.

[0199] like Figure 5 FIG. 1 is a schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 5 As shown, the electronic device 10 of this embodiment includes: a processor 100, a memory 101, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, the steps of the above-described method embodiment are implemented. Alternatively, when the processor 100 executes the computer program 102, the functions of the modules / units in the above-described device embodiment are implemented.

[0200] Exemplarily, the computer program 102 can be divided into one or more modules / units, one or more of which are stored in the memory 101 and executed by the processor 100 to implement the present invention. One or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program 102 in the electronic device 10. For example, the computer program 102 can be divided into a deviation sequence acquisition module, a control deviation oscillation rate calculation module, a valve output deviation oscillation rate calculation module, and a valve sticking risk determination module.

[0201] The electronic device 10 can be a computing device such as a desktop computer, a notebook computer, a palm computer, a cloud server, etc. The electronic device 10 can include, but is not limited to, a processor 100, a memory 101. Those skilled in the art can understand that Figure 5 The electronic device 10 is merely an example and does not constitute a limitation on the electronic device 10, and can include more or fewer components than shown, or combine certain components, or different components, for example, the electronic device can also include an input / output device, a network access device, a bus, etc.

[0202] The processor 100 can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0203] The memory 101 can be an internal storage unit of the electronic device 10, such as a hard disk or a memory of the electronic device 10. The memory 101 can also be an external storage device of the electronic device 10, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 101 can include both an internal storage unit and an external storage device of the electronic device 10. The memory 101 is used to store computer programs and other programs and data required by the electronic device 10. The memory 101 can also be used to temporarily store data that has been output or will be output.

[0204] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is taken as an example, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction, and do not limit the protection scope of the application. The specific working process of the units and modules in the system can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0205] Those skilled in the art will appreciate that embodiments of the application can be provided as methods, systems, or computer program 102 products. Therefore, the application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the application can take the form of a computer program 102 product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0206] The application is described with reference to flowcharts and / or block diagrams according to the method, device (system), and computer program 102 product of the embodiments of the application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram can be implemented by computer program 102 instructions. These computer program 102 instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowchart and / or block diagram. Figure 1 one flow or multiple flows and / or blocks Figure 1 an apparatus that performs the functions specified in one block or multiple blocks.

[0207] These computer program 102 instructions can also be stored in a computer-readable storage 101 that can guide the computer or other programmable data processing apparatus to work in a specific way, so that the instructions stored in the computer-readable storage 101 produce a manufactured product including instruction apparatus, which implements the functions specified in the flowchart and / or block diagram. Figure 1 one flow or multiple flows and / or blocks Figure 1 an apparatus that performs the functions specified in one block or multiple blocks.

[0208] These computer program instructions 102 can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 Figure 1

[0209] It is also important to note that the terms "comprises" and / or "comprising", or "includes" and / or "including" when used in this specification, specify the presence of stated features, integers, steps, operations, elements, or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or groups thereof.

[0210] The embodiments of method, device, and system of the present application can take many different forms. The present application should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided for illustrative purposes. Although the present application has been described in some detail, those skilled in the art should appreciate that many modifications are possible that will come within the scope of the present application as defined by the appended claims. Accordingly, the terms of such claims should not be construed as limiting the present application to the precise steps or other such tightly-defined formulations.​​

Claims

1. A valve viscosity identification method, characterized in that: include: Based on the historical operating data of the target control loop, a control deviation sequence and a valve output deviation sequence are obtained; Based on the control deviation sequence, the control deviation oscillation rate and the control deviation abnormal waveform oscillation rate are calculated; Based on the valve output deviation sequence, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate are calculated; Based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate, the risk judgment of valve sticking is performed to obtain a risk judgment result.

2. The valve viscosity identification method according to claim 1, characterized in that: The historical operation data of the target control loop includes the set value sequence SP[I], the measurement value sequence PV[I] and the output value sequence OP[I] of the target control loop, where I={1, 2, ..., n}, and n is the number of data collected; The process of obtaining a control deviation sequence and a valve output deviation sequence based on historical operation data of the target control loop includes: According to the deviation between the measured value sequence PV[I] and the set value sequence SP[I], the control deviation sequence e[I] is obtained; wherein e[I]=PV[I]-SP[I]; According to the deviation between the output value sequence OP[I] and the average output value of the output value sequence OP[I], the valve output deviation sequence OPe[I] is obtained; wherein, 3. The valve viscosity identification method according to claim 1, characterized in that: The control deviation oscillation rate and the control deviation abnormal waveform oscillation rate are calculated based on the control deviation sequence, including: Classifying and assigning values ​​to the control deviation sequence to obtain a plurality of first array data sequences; Based on the minimum distance method, the average value and the average value similarity are calculated for each first array data sequence respectively; Based on the average value and average value similarity of each first array data sequence, abnormal waveform oscillation is identified, and the control deviation oscillation rate and the oscillation rate of each abnormal waveform are calculated; wherein the abnormal waveform includes a square waveform and a triangular waveform; Based on the oscillation rate of each abnormal waveform, the control deviation abnormal waveform oscillation rate is obtained.

4. The valve viscosity identification method according to claim 3, characterized in that: The control deviation sequence is classified and assigned values ​​to obtain a plurality of first array data sequences, including: Traversing the control deviation sequence, and filling the pre-established array corresponding to each first array data sequence based on the positive and negative change information of the data in the control deviation sequence, to obtain multiple first array data sequences; Among them, the multiple first array data sequences include continuous positive deviation sequences, continuous positive sum sequences, continuous positive duration sequences, continuous positive maximum time point sequences, continuous negative deviation sequences, continuous negative sum sequences, continuous negative duration sequences and continuous negative maximum time point sequences.

5. The valve viscosity identification method according to claim 4, characterized in that: The method of filling the pre-established array corresponding to each first array data sequence based on the positive and negative change information of the data of the control deviation sequence to obtain multiple first array data sequences includes: When the data of the first subsequence of the control deviation sequence are continuously positive and the data adjacent to the last data of the first subsequence are negative, the data of the first subsequence are stored in a continuous positive deviation sequence, the positive sum of the data of the first subsequence is stored in a continuous positive sum sequence, the acquisition time points corresponding to the data of the first subsequence are stored in a continuous positive duration sequence, and the acquisition time points in the first subsequence that meet the positive maximum time point are stored in a continuous positive maximum time point sequence; wherein the positive maximum time point represents all acquisition time points from the acquisition time point of the first data of the first subsequence to the acquisition time point corresponding to the maximum value of the data in the first subsequence; When the data of the first subsequence of the control deviation sequence are continuously negative and the data adjacent to the end data of the first subsequence are positive, the data of the first subsequence are stored in a continuous negative deviation sequence, the negative sum of the data of the first subsequence is stored in a continuous negative sum sequence, the acquisition time points corresponding to the data of the first subsequence are stored in a continuous negative duration sequence, and the acquisition time points in the first subsequence that meet the negative maximum time point are stored in a continuous negative maximum time point sequence; wherein the negative maximum time point represents all acquisition time points from the acquisition time point of the first data of the first subsequence to the acquisition time point corresponding to the negative maximum value of the data in the first subsequence; wherein, The control deviation sequence is e[I], I={1, 2, ..., n}, and the first subsequence is e[i], i={a, a+1, ..., b}, a≥1, b is less than n.

6. The valve viscosity identification method according to claim 3, characterized in that: The method of calculating the average value and the average value similarity of each first array data sequence based on the minimum distance method includes: According to the formula Determine the average value K[imin] corresponding to each first array data sequence; where K[i] represents the data corresponding to time i in the first array data sequence, K[j] represents the data corresponding to time j in the first array data sequence, K[imin] represents the data corresponding to time imin in the first array data sequence, and imin is The moment corresponding to the minimum value of The sum of the squares of the deviations between all data in the first array data sequence and K[imin] does not exceed the sum of the squares of the deviations between all data in the first array data sequence and K[j]; where j = 1, 2, ..., m, i = 1, 2, ..., m, and m is greater than 10; For each first array data sequence, determine and remove data in the first array data sequence that is not close to the corresponding average value K[imin], and obtain a first new array data sequence Knew[I] that is close to the average value K[imin]; where I = {1, 2, ..., z}, z ≤ m; Based on each first new array data sequence, the corresponding latest average value Avgnew and the average value similarity are calculated.

7. The valve viscosity identification method according to claim 6, characterized in that: The determining and removing data in the first array data sequence that is not close to the corresponding average value K[imin] to obtain a first new array data sequence Knew[I] that is close to the average value K[imin] includes: When the average value K[imin] is zero, set the parameter z to zero so that the corresponding first new array data sequence Knew[I] is an empty sequence; when or When , it means that the K[i] value is not close to the corresponding average value K[imin], and the data K[i] at time i in the first array data sequence is cleared to obtain the first new array data sequence Knew[I] close to the average value K[imin]; wherein, When the parameter z is less than a preset value, the corresponding average similarity is determined to be zero, i=1, 2, . . . , m.

8. The valve viscosity identification method according to claim 6, characterized in that: The calculation formula of the average similarity is as follows: Among them, si represents the average similarity, and Avgnew represents the latest average value.

9. The valve viscosity identification method according to claim 3, characterized in that: The identifying of abnormal waveform oscillation based on the average value and average value similarity of each first array of data sequences includes: Combining the average values ​​and average similarities of the first array data sequences to obtain a corresponding combined judgment set; Based on a preset waveform oscillation determination rule library, matching the waveform oscillation determination result corresponding to the combination determination set; wherein, The preset waveform oscillation determination rule library stores determination rules corresponding to all waveform oscillation determination results.

10. The valve viscosity identification method according to claim 3, characterized in that: The first data set includes a continuous positive deviation sequence and a continuous negative deviation sequence; The oscillation rate of the square waveform is calculated as follows: Where FRateF represents the oscillation rate of the square waveform, fASOsc represents the average similarity of the continuous positive deviation sequence, and fBSOsc represents the average similarity of the continuous negative deviation sequence.

11. The valve viscosity identification method according to claim 3, characterized in that: The first array data sequence includes a sequence of continuous positive maximum time points and a sequence of continuous negative maximum time points; The oscillation rate of the triangular waveform is calculated as follows: Where FRateS represents the oscillation rate of the triangular waveform, AMTavg represents the average value of the sequence of consecutive positive maximum values, and BMTavg represents the average value of the sequence of consecutive negative maximum values.

12. The valve viscosity identification method according to claim 3, characterized in that: The method of obtaining the control deviation abnormal waveform oscillation rate based on the oscillation rate of each abnormal waveform includes: The maximum value of the oscillation rates of all abnormal waveforms is taken as the oscillation rate of the abnormal waveform of the control deviation; eFRate=max(FRateF, FRateS), where eFRate represents the oscillation rate of the abnormal control deviation waveform, FRateF represents the oscillation rate of the square waveform, and FRateS represents the oscillation rate of the triangle waveform.

13. The valve viscosity identification method according to claim 3, characterized in that: The first array data sequence includes a continuous positive sum sequence, a continuous positive duration sequence, a continuous negative sum sequence and a continuous negative duration sequence; The calculation formula of the control deviation oscillation rate is as follows: Among them, eRate represents the control deviation oscillation rate, ASOsc represents the average similarity rate of consecutive positive sum sequences, ATOsc represents the average similarity rate of consecutive positive duration sequences, BSOsc represents the average similarity rate of consecutive negative sum sequences, and BTOsc represents the average similarity rate of consecutive negative duration sequences.

14. The valve viscosity identification method according to claim 1, characterized in that: The valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate are calculated based on the valve output deviation sequence, including: Classifying and assigning values ​​to the valve output deviation sequence to obtain a plurality of second array data sequences; Based on the minimum distance method, calculating the average value and the average value similarity of each second array data sequence respectively; Based on the average value and average value similarity of each second array data sequence, abnormal waveform oscillation is identified, and the valve output deviation oscillation rate and the oscillation rate of each abnormal waveform are calculated; wherein the abnormal waveform includes a square waveform and a triangular waveform; Based on the oscillation rate of each abnormal waveform, the oscillation rate of the valve output deviation abnormal waveform is obtained.

15. The valve sticking identification method according to claim 1, characterized in that: The risk assessment of valve sticking based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate, and the valve output deviation abnormal waveform oscillation rate includes: The valve viscosity rate St is obtained based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate; The risk of valve sticking is determined based on the valve sticking rate St. The closer the valve sticking rate St is to 100%, the higher the positivity of the valve sticking is. When the valve sticking rate St is greater than 80%, a valve sticking risk signal is generated.

16. The valve sticking identification method according to claim 15, characterized in that: The calculation formula of the valve viscosity St is as follows: Wherein, eRate represents the control deviation oscillation rate, eFRate represents the control deviation abnormal waveform oscillation rate, OPRate represents the valve output deviation oscillation rate, and OPFRate represents the valve output deviation abnormal waveform oscillation rate.

17. The valve sticking identification method according to claim 1, characterized in that: Also includes: Obtain historical operating data of the target control loop through the industrial control data acquisition interface and / or database interface.

18. A valve viscosity identification system, characterized in that: include: A deviation sequence obtaining module is used to obtain a control deviation sequence and a valve output deviation sequence based on historical operation data of a target control loop; A control deviation oscillation rate calculation module is used to calculate the control deviation oscillation rate and the control deviation abnormal waveform oscillation rate based on the control deviation sequence; A valve output deviation oscillation rate calculation module is used to calculate the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate based on the valve output deviation sequence; The valve sticking risk judgment module is used to judge the risk of valve sticking based on the control deviation oscillation rate, the control deviation abnormal waveform oscillation rate, the valve output deviation oscillation rate and the valve output deviation abnormal waveform oscillation rate to obtain a risk judgment result.

19. A machine-readable storage medium having instructions stored thereon, characterized in that: When the instruction is executed by a processor, the processor is configured to execute the valve sticking identification method according to any one of claims 1 to 17.

20. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the valve viscosity identification method according to any one of claims 1 to 17 is implemented.

Images