Thermal power generating unit control performance evaluation method and device and electronic equipment

Abstract

The invention provides a thermal power generating unit control performance evaluation method and device and electronic equipment, and belongs to the technical field of thermal power generating units. The method comprises the following steps: acquiring distributed control system historical data of a thermal power generating unit in a preset time period; state recognition is carried out on historical data of the distributed control system based on the data change trend, and a dynamic adjustment process data set and a steady-state random process data set are obtained; a deterministic performance evaluation index is calculated based on the dynamic adjustment process dataset, and a randomness performance evaluation index is calculated based on the steady-state random process dataset. Compared with the method for measuring the control performance of the thermal power generating unit by directly adopting indexes such as adjusting time, rising time and overshoot, the method is easily influenced by large disturbance conditions, and the accuracy of control performance evaluation is reduced. According to the method, the historical data of the distributed control system is divided into two parts of data to calculate the control performance, the influence of large disturbance conditions is reduced, and therefore the accuracy of control performance evaluation of the thermal power generating unit is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal power units, and in particular to a thermal power unit control performance evaluation method, a thermal power unit performance evaluation device and an electronic device. Background Art

[0002] In large thermal power units, the control loop is an important part of the unit operation. The regulation quality of the control loop can reflect the operating status of the thermal power unit. In the long-term operation of the thermal power unit, the control performance of the control loop will inevitably decline. Therefore, it is necessary to read the main automatic data of the machine side and the furnace side of the thermal power unit during operation, evaluate the control performance, and return the evaluation results to the distributed control system (DCS for short) for reference by the operators.

[0003] For thermal power units, the control performance is usually measured by indicators such as adjustment time, rise time and overshoot, but these indicators require a step response and can only be obtained when there is no large disturbance. In the actual production process, it is difficult to obtain good analysis data due to complex on-site conditions, or the good data obtained is very limited, which reduces the accuracy of the control performance evaluation of thermal power units. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a method, device and electronic equipment for evaluating the control performance of a thermal power unit, so as to solve the defect of low accuracy of the performance evaluation of the thermal power unit in the existing solutions.

[0005] In order to achieve the above object, an embodiment of the present invention provides a method for evaluating control performance of a thermal power unit, the method comprising:

[0006] Obtaining historical data of the distributed control system of the thermal power unit within a preset time period;

[0007] Based on the data change trend, the state of the historical data of the distributed control system is identified to obtain a dynamic adjustment process data set and a steady-state random process data set;

[0008] Calculating a deterministic performance evaluation index based on the dynamic adjustment process data set, and calculating a stochastic performance evaluation index based on the steady-state random process data set;

[0009] Among them, the data change trend of the dynamic adjustment process data set is greater than the data change trend of the steady-state random process data set; the deterministic performance evaluation index characterizes the dynamic adjustment performance of the thermal power unit, and the random performance evaluation index characterizes the performance of the thermal power unit in resisting random disturbances in the steady-state process.

[0010] Optionally, the state identification of the historical data of the distributed control system based on the data change trend to obtain a dynamic adjustment process data set and a steady-state random process data set includes:

[0011] Performing piecewise linearization processing on the historical data of the distributed control system to obtain piecewise linearization data;

[0012] Based on the data variation trend of each straight line segment in the piecewise linearization data, the state of the piecewise linearization data is identified to obtain a dynamic adjustment process data set and a steady-state random process data set.

[0013] Optionally, the state recognition of the piecewise linearized data is performed based on the data change trend of each straight line segment in the piecewise linearized data to obtain a dynamic adjustment process data set and a steady-state random process data set, including:

[0014] Calculating the slope of each straight line in the piecewise linearization data;

[0015] Determine the change in the slope of each straight line of the piecewise linearization data based on the slope of each straight line;

[0016] The change in the slope of the straight line is identified based on a set threshold, and a dynamic adjustment process data set and a steady-state random process data set are obtained.

[0017] Optionally, the calculating of a randomness performance evaluation index based on the steady-state random process data set includes:

[0018] Performing time series analysis based on the steady-state random process data set and establishing an autoregressive model;

[0019] Determining the theoretical minimum variance of a steady-state random process data set based on the autoregressive model;

[0020] Based on the theoretical minimum variance and the actual output variance of the steady-state random process data set, the minimum variance is calculated to obtain the randomness performance evaluation index.

[0021] Optionally, the calculating of a deterministic performance evaluation index based on the dynamic adjustment process data set includes:

[0022] Constructing a linear prediction model based on the dynamic adjustment process data set;

[0023] Estimating model parameters of the linear prediction model based on a gradient descent method to obtain estimated values;

[0024] Performing residual check on the linear prediction model based on the estimated value;

[0025] Based on the residual-checked linear prediction model, the rise time, the adjustment time and the overshoot are calculated to serve as the deterministic performance evaluation indicators.

[0026] Optionally, the distributed control system historical data includes sensor historical data; after acquiring the distributed control system historical data of the thermal power unit within a preset time period, the method further includes:

[0027] A fault analysis is performed on the rate of change of the value of the historical data of the sensor based on a set rate of change threshold.

[0028] Optionally, the distributed control system historical data includes historical opening instructions of the actuator; after acquiring the distributed control system historical data of the thermal power unit within a preset time period, the method further includes:

[0029] A fault analysis is performed on the actuator based on the historical opening instructions and the opening feedback corresponding to the historical opening instructions.

[0030] On the other hand, an embodiment of the present invention further provides a thermal power unit control performance evaluation device, comprising:

[0031] A data acquisition module is used to acquire the historical data of the distributed control system of the thermal power unit within a preset time period;

[0032] A state identification module, used to identify the state of the historical data of the distributed control system based on the data change trend, and obtain a dynamic adjustment process data set and a steady-state random process data set;

[0033] A performance evaluation module, used to calculate a deterministic performance evaluation index based on the dynamic adjustment process data set, and to calculate a stochastic performance evaluation index based on the steady-state random process data set;

[0034] Among them, the data change trend of the dynamic adjustment process data set is greater than the data change trend of the steady-state random process data set; the deterministic performance evaluation index characterizes the dynamic adjustment performance of the thermal power unit, and the random performance evaluation index characterizes the performance of the thermal power unit in resisting random disturbances in the steady-state process.

[0035] On the other hand, the present invention also provides 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 program, a control performance evaluation method for a thermal power unit as described above is implemented.

[0036] On the other hand, the present invention further provides a machine-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for evaluating the control performance of a thermal power unit as described in any one of the above is implemented.

[0037] On the other hand, the present invention further provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for evaluating the control performance of a thermal power unit.

[0038] Through the above technical scheme, compared with the existing scheme that directly uses indicators such as adjustment time, rise time and overshoot to measure the control performance of thermal power units, it is easy to be affected by large disturbances, resulting in reduced accuracy of control performance evaluation. The present invention divides the historical data of the distributed control system into a dynamic adjustment process data set in a dynamic adjustment process and a steady-state random process data set in a steady-state random process according to the data change trend, and then calculates the deterministic performance evaluation index in the dynamic adjustment process and the random performance evaluation index in the steady-state random process. By dividing the historical data of the distributed control system into two parts of data to calculate the control performance, the influence of large disturbances is reduced, thereby improving the accuracy of the control performance evaluation of the thermal power unit.

[0039] 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

[0040] 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:

[0041] Figure 1 This is one of the flow diagrams of the thermal power unit control performance evaluation method provided by the present invention;

[0042] Figure 2 It is a schematic diagram for comparing the original data and the segmented linearized data provided by the present invention;

[0043] Figure 3 is a schematic diagram of a single variable closed-loop feedback control system provided by the present invention;

[0044] Figure 4 This is the second flow chart of the thermal power unit control performance evaluation method provided by the present invention;

[0045] Figure 5 This is the third flow chart of the thermal power unit control performance evaluation method provided by the present invention;

[0046] Figure 6 This is the fourth flow chart of the thermal power unit control performance evaluation method provided by the present invention;

[0047] Figure 7 It is a structural schematic diagram of a thermal power unit control performance evaluation device provided by the present invention;

[0048] Figure 8 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0049] 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.

[0050] Method Embodiment

[0051] Please refer to Figure 1 The embodiment of the present invention provides a method for evaluating the control performance of a thermal power unit, the method comprising:

[0052] Step 100: Obtain historical data of the distributed control system of the thermal power unit within a preset time period.

[0053] The electronic device obtains the historical data of the distributed control system of the thermal power unit within a preset time period. Specifically, the electronic device can use 1s as a sampling period to obtain the historical data of the distributed control system of the thermal power unit within a preset time period (for example, within one hour). The historical data of the distributed control system may include various data of the thermal power unit, such as sensor data and actuator opening instructions.

[0054] It should be noted that the method of the embodiment of the present invention can be used for the control performance evaluation of a 660MW single-column auxiliary thermal power unit.

[0055] Step 200: Perform state recognition on the historical data of the distributed control system based on the data change trend to obtain a dynamic adjustment process data set and a steady-state random process data set.

[0056] The electronic device performs state recognition on the historical data of the distributed control system based on the data change trend to obtain a dynamic adjustment process data set and a steady-state random process data set, wherein the data change trend of the dynamic adjustment process data set is greater than the data change trend of the steady-state random process data set.

[0057] By dividing the historical data of the distributed control system into a dynamic regulation process data set in a dynamic regulation process and a steady-state random process data set in a steady-state random process, it is beneficial to calculate the control performance in the dynamic regulation process and the steady-state random process respectively.

[0058] Step 300: Calculate a deterministic performance evaluation index based on the dynamic adjustment process data set, and calculate a random performance evaluation index based on the steady-state random process data set.

[0059] The electronic device calculates a deterministic performance evaluation index based on the dynamic regulation process data set, and calculates a random performance evaluation index based on the steady-state random process data set. The deterministic performance evaluation index characterizes the dynamic regulation performance of the thermal power unit, and the random performance evaluation index characterizes the performance of the thermal power unit in resisting random disturbances in a steady-state process.

[0060] Specifically, for the dynamic adjustment process data set, the existing performance indicators can be used, that is, the control performance can be evaluated by the adjustment time, rise time and overshoot of the dynamic adjustment process data set. Combined with the mathematical model of the dynamic adjustment process data set identification, a step disturbance is added to the mathematical model to obtain the deterministic performance evaluation index of the entire adjustment process.

[0061] For steady-state random process data sets, the minimum variance index (i.e., randomness performance evaluation index) obtained based on the theoretical minimum variance and the actual output variance can be used to evaluate the ability of thermal power units to resist random disturbances. The minimum variance value range is [0, 1]. The closer it is to 0, the worse the control performance of the thermal power unit in the sense of minimum variance. The closer it is to 1, the better the control performance is.

[0062] Compared with the existing scheme that directly uses the adjustment time, rise time and overshoot and other indicators to measure the control performance of the thermal power unit, it is easily affected by large disturbances, resulting in reduced accuracy of control performance evaluation. The present invention divides the historical data of the distributed control system into a dynamic adjustment process data set in a dynamic adjustment process and a steady-state random process data set in a steady-state random process according to the data change trend, and then calculates the deterministic performance evaluation index in the dynamic adjustment process and the random performance evaluation index in the steady-state random process. By dividing the historical data of the distributed control system into two parts of data to calculate the control performance, the influence of large disturbances is reduced, thereby improving the accuracy of the control performance evaluation of the thermal power unit.

[0063] In other aspects of the embodiments of the present invention, step 200, performing state recognition on the historical data of the distributed control system based on the data change trend to obtain a dynamic adjustment process data set and a steady-state random process data set, includes:

[0064] Step 210: Perform piecewise linearization processing on the historical data of the distributed control system to obtain piecewise linearization data.

[0065] Please refer to Figure 2 The electronic device performs piecewise linearization processing on the historical data of the distributed control system to obtain piecewise linearization data. Figure 2 The horizontal axis is time in seconds, and the vertical axis is the value. The piecewise linearization process converts the original data (ie, the historical data of the distributed control system) into piecewise linearized data composed of different line segments.

[0066] Step 220: Based on the data variation trend of each straight line segment in the piecewise linearization data, state recognition is performed on the piecewise linearization data to obtain a dynamic adjustment process data set and a steady-state random process data set.

[0067] The electronic device performs state recognition on the piecewise linearization data based on the data change trend of each straight line in the piecewise linearization data to obtain a dynamic adjustment process data set and a steady-state random process data set. For example, a plurality of data with a large data change trend of the straight line in the piecewise linearization data is determined as a dynamic adjustment process data set, and a plurality of data with a small data change trend of the straight line in the piecewise linearization data is determined as a steady-state random process data set.

[0068] In other aspects of the embodiments of the present invention, step 220, based on the data change trend of each straight line segment in the piecewise linearization data, performs state recognition on the piecewise linearization data to obtain a dynamic adjustment process data set and a steady-state random process data set, includes:

[0069] Step 221: Calculate the slope of each straight line segment in the piecewise linearization data.

[0070] The electronics calculate the line slope for each line segment in the piecewise linearization data.

[0071] Step 222: Determine the change in the slopes of adjacent straight lines in the piecewise linearization data based on the slopes of the straight lines of the straight lines.

[0072] The electronic device determines the change in the slope of the adjacent straight lines in the piecewise linearization data based on the slope of each straight line segment. For example, the slope of straight line a in the piecewise linearization data is 2, and the slope of straight line b adjacent to straight line a is 0. Then the change in the slope of straight line a and straight line b is |-2|=2.

[0073] Step 223: Identify the change in the slope of the straight line based on a set threshold to obtain a dynamic adjustment process data set and a steady-state random process data set.

[0074] The electronic device identifies the change in the slope of the straight line based on the set threshold value, and obtains a dynamic adjustment process data set and a steady-state random process data set. For example, the set threshold value is set to 5. When the change in the slope of the straight line is less than or equal to 5, such as the change in the slope of the straight line of straight line a and straight line b is 2, the data corresponding to straight line a and straight line b are steady-state random process data sets. For another example, when the change in the slope of the straight line is greater than 5, such as the change in the slope of the adjacent straight line c and straight line d is 6, the data corresponding to straight line c and straight line d are dynamic adjustment process data sets. Therefore, the embodiment of the present invention identifies each segment of the piecewise linearized data by setting a threshold value, and obtains a dynamic adjustment process data set and a steady-state random process data set. So as to subsequently calculate the deterministic performance evaluation index of the dynamic adjustment process, and calculate the random performance evaluation index of the steady-state random process.

[0075] In other aspects of the embodiments of the present invention, step 300, calculating a randomness performance evaluation index based on the steady-state random process data set, includes:

[0076] Step 310: Perform time series analysis based on the steady-state random process data set to establish an autoregressive model.

[0077] The electronic device performs time series analysis based on the steady-state random process data set to establish an autoregressive model. The embodiment of the present invention considers a single variable closed-loop feedback control system (i.e., an autoregressive model) such as Figure 3 As shown. Where C is the transfer function of the controller, D is the disturbance transfer function, P is the transfer function of the controlled object and d is the process delay, is a pure delay-free transfer function, ε(t) has a mean of 0 and a variance of of white noise signal.

[0078] The closed-loop feedback control system (i.e., autoregressive model) can be described as:

[0079]

[0080] in:

[0081]

[0082]

[0083] in As a differential operator, the disturbance transfer function D is decomposed according to the Diophantine equation:

[0084]

[0085] where f i(i=1,2,...,d-1) is a fixed coefficient, R=f d +f d+1 z -1 +...is a rational canonical transfer function. Assuming the setting value r(t)=0, then

[0086]

[0087]

[0088] here is a regular transfer function, so the two parts on the right side of the above equation are independent of each other, then:

[0089]

[0090] Step 320: Determine the theoretical minimum variance of the steady-state random process data set based on the autoregressive model.

[0091] The above formula can obtain the theoretical minimum variance The condition is The corresponding minimum variance controller can be obtained as:

[0092]

[0093] The transfer function C of the Fε(t) controller is independent and is called the feedback control invariant.

[0094] Step 330: Calculate the minimum variance based on the theoretical minimum variance and the actual output variance of the steady-state random process data set to obtain the randomness performance evaluation index.

[0095] The electronic equipment calculates the actual output variance of the thermal power unit from the conventional operation data, and then the minimum variance is obtained by ratioing the theoretical minimum variance with the actual output variance. The minimum variance is determined as the randomness performance evaluation indicator.

[0096] Specifically, for a steady-state process, its output can be expressed as an infinite moving average model, and the theoretical minimum variance is estimated using the first d terms Fε(t) of the model. Using the minimum variance as the benchmark value, the minimum variance η can be obtained Harris

[0097]

[0098] η Harris It can be understood as the theoretical minimum variance of the system due to disturbance within the time lag The actual output variance The ratio of η to η is η, because the actual output variance cannot be smaller than the minimum variance in the ideal case, so ηHarris The value is between 0 and 1. The closer it is to 1, the better the actual thermal power unit control performance is. The closer it is to 0, the worse the current thermal power unit controller performance is. There is still a lot of room for improvement from the ideal situation.

[0099] In other aspects of the embodiments of the present invention, step 300, calculating a deterministic performance evaluation index based on the dynamic adjustment process data set, includes:

[0100] Step 350: construct a linear prediction model based on the dynamic adjustment process data set.

[0101] Specifically, the electronic device can use the dynamic adjustment process data set to build a linear prediction model. For example, the embodiment of the present invention can build an ARMAX prediction model. The ARMAX prediction model defines the time series as a multi-input multi-output type, which can be expressed as follows:

[0102] y(t)+a1y(t-1)+a2y(t-2)+...+a na y(t-na)

[0103] =b1u(t-1)+b2u(t-2)+...+b nb u(t-nb)+e(t)+c1e(t-1)+c2e(t-2)+...+c n e(t-nc)

[0104] Where n = max{na,nb,nc}, na is the order of the AR model, nb is the order of the input model, nc is the order of the MA model, u(t) represents the input of the system at time t, y(t) represents the output of the system at time t, e(t) represents the random noise term, and the model's parameters to be identified are represented by θ, then:

[0105] θ=[a1,...,a na ,b1,...,b nb ,c1,...,c nc ]

[0106] The ARMAX forecasting model can be abbreviated as:

[0107] A(q)y(t)=B(q)u(t)+C(q)e(t)

[0108] The polynomials are expressed as:

[0109] A(q)=1+a1q -1 +a2q -2 +...+a na q -na

[0110] B(q)=b1q-1 +b2q -2 +...+b nb q -nb

[0111] C(q)=1+c1q -1 +c2q -2 +...+c nc q -nc

[0112] where q -1 is a backward shift operator.

[0113] Step 360: estimate the model parameters of the linear prediction model based on the gradient descent method to obtain estimated values.

[0114] Specifically, the electronic device can obtain the estimated value of the model parameter θ by minimizing the ARMAX prediction model of the above formula through the gradient descent method

[0115]

[0116] Step 370: Perform residual check on the linear prediction model based on the estimated value.

[0117] The electronic device obtains the estimated value of the model parameter θ Then, the estimated residual autocorrelation coefficient is calculated according to the following formula: According to Perform a residual test with a certain confidence level to test whether the ARMAX prediction model passes the residual test.

[0118]

[0119] It should be noted that step 350 to step 370 can be understood as the model identification stage of the linear prediction model.

[0120] Step 380: Based on the linear prediction model that has been verified by residual errors, the rise time, the adjustment time and the overshoot are calculated to serve as the deterministic performance evaluation indicators.

[0121] After residual verification, the ARMAX prediction model of the thermal power unit control loop is obtained, and then the rise time, adjustment time and overshoot are calculated as the deterministic performance evaluation indicators. The structure diagram of a typical single variable closed-loop feedback control system is shown in Figure 2. Figure 3 The main steps are as follows:

[0122] S1. Calculate the closed-loop transfer function of the ARMAX prediction model of thermal power units:

[0123]

[0124]

[0125] Among them, G is the closed-loop transfer function; P is the controlled object transfer function; C is the controller transfer function.

[0126] S2. Obtain the step response curve based on the closed-loop transfer function, and obtain the zero-crossing time t0, rise time, and adjustment time t s , overshoot and other deterministic performance evaluation indicators.

[0127] S3. Compare the obtained deterministic performance evaluation index with the preset control performance of the thermal power unit, and adjust the parameters of the target controller according to the following PID optimization rules:

[0128] If t0, t s If the preset upper limit of the corresponding thermal power unit is exceeded, the proportional action or integral action is increased;

[0129] If t0, t s If it is lower than the preset lower limit of the corresponding thermal power unit and the number of oscillations exceeds the preset value of the thermal power unit, the proportional action or integral action is reduced.

[0130] It should be noted that, before step 350, building a linear prediction model based on the dynamic adjustment process data set, the method further includes step 340, performing data preprocessing on the dynamic adjustment process data set.

[0131] Step 340: preprocessing the dynamic adjustment process data set, specifically including:

[0132] The dynamic adjustment process data set may include historical data of a controlled variable, a set value, and an operating variable of a control loop of the thermal power unit for a preset period of time.

[0133] Step 341: Remove outliers from the dynamic adjustment process data set. Before removing outliers from the dynamic adjustment process data set, first use the following formula to calculate the standard deviation of the controlled variable and the manipulated variable:

[0134]

[0135]

[0136] The controlled variable PV obtained for each sampling is expressed as y k , the set value is denoted as r k , the manipulated variable MV is represented by u k , where the subscript k is the sampling time, and after sampling N times, we get y∈R 1×N ,r∈R 1×N ,u∈R 1×N

[0137] When |y k -mean(y k )|>3σ y or |u k -mean(u k )|>3σ u When , the value corresponding to the k sampling moment is eliminated;

[0138] Step 342: Determine whether oscillation occurs according to the oscillation intensity index reg, and filter out the non-oscillating operation data of the dynamic adjustment process data set. Use the autocorrelation function to establish an oscillation intensity index to indicate whether oscillation occurs. When the oscillation intensity index exceeds the preset threshold, it is considered that oscillation occurs in the loop during this time period. The autocorrelation function is as follows:

[0139]

[0140] x t =y t -r t

[0141] Among them, x t is the deviation sequence, σ is the sequence x t The standard deviation of the sequence x t The mean value, ρ τ represents the autocorrelation coefficient at time lag τ.

[0142] With τ as the horizontal axis, ρ τ As the vertical coordinate, we can get the value of (τ, ρ τ ) function graph, record the image ρ τ The zero crossing interval is variable T p , construct the oscillation strength indicator:

[0143]

[0144]

[0145] in, is the mean of the zero-crossing intervals, ΔT p is a random variable, For the variable T p The standard deviation of the distribution, reg is an indicator of the strength of the oscillation.

[0146] Step 343: Screen the identification data of the dynamic adjustment process data set according to the Fisher information matrix condition number. The screening rule is to require the condition number value of the data segment to be small and stable. Construct the Fisher information matrix, determine the average amount of information about the identification parameters that the manipulated variables and the controlled variables can provide, and screen the data for identification according to the matrix condition number.

[0147] The variables for constructing the Fisher information matrix are as follows:

[0148] y t =y t-1 α+u t-1 β

[0149] y t-1 =(y t-1 ,y t-2 ,...,y t-na )

[0150] u t-1 =(u t-1 ,u t-2 ,...,u t-nb )

[0151] α=(α1,α2,...,α na ) T

[0152] β=(β1,β2,...,β nb ) T

[0153] Among them, y t ,u t They represent the controlled variable sequence and the control variable sequence starting at time t, α, β are the parameter vectors to be identified of the model, na, nb are the dimensions of the parameter vectors α, β respectively.

[0154] The constructed Fisher information matrix is ​​expressed as follows:

[0155]

[0156] Among them, F represents the information matrix and E() represents the expectation.

[0157] The condition number of the Fisher information matrix is ​​calculated according to the following formula:

[0158]

[0159] Among them, λ(F) is the eigenvalue of matrix F. The historical data whose condition number of Fisher information matrix is ​​less than the threshold are selected as identification data.

[0160] Through the above steps, high-quality identification data that meets the identification requirements of thermal power units can be obtained to build a linear prediction model.

[0161] For other aspects of the embodiments of the present invention, please refer to Figure 4, the distributed control system historical data includes sensor historical data; after step 100, obtaining the distributed control system historical data of the thermal power unit within a preset time period, it also includes:

[0162] Step 400: Perform fault analysis on the rate of change of the value of the historical data of the sensor based on a set rate of change threshold.

[0163] Specifically, a preset change rate threshold is set in the electronic device to measure the rate of change of the value of the sensor's historical data, and then perform fault analysis. For example, when the rate of change of the value of the sensor's historical data is greater than the set change rate threshold, the electronic device determines that the sensor has a change rate exceeding limit fault. When the rate of change of the value of the sensor's historical data is less than or equal to the set change rate threshold, the electronic device determines that the sensor does not have a change rate exceeding limit fault. Thus, the embodiment of the present invention realizes fault analysis of the sensor's historical data. It provides an effective basis for further control optimization, efficiency improvement, and energy conservation and emission reduction.

[0164] It should be noted that a fault threshold can also be preset in the electronic device to measure the sensor historical data for fault analysis. For example, when the value of the sensor historical data is greater than the preset fault threshold, the electronic device determines that the sensor value has an over-limit fault. When the value of the sensor historical data is less than or equal to the preset fault threshold, the electronic device determines that the sensor does not have an over-limit fault. Thus, the embodiment of the present invention realizes fault analysis of the sensor historical data. It provides an effective basis for further control optimization, efficiency improvement, and energy conservation and emission reduction.

[0165] For other aspects of the embodiments of the present invention, please refer to Figure 5 , the distributed control system historical data includes the historical opening instructions of the actuator; after step 100, obtaining the distributed control system historical data of the thermal power unit within a preset time period, it also includes:

[0166] Step 500: Perform fault analysis on the actuator based on the historical opening instructions and the opening feedback corresponding to the historical opening instructions.

[0167] Specifically, within a certain period of time, when the historical opening instruction of the actuator changes, but the opening feedback does not change with the historical opening instruction of the actuator, or when the historical opening instruction of the actuator does not change, the opening feedback does not change, but there is a deviation between the opening feedback and the historical opening instruction of the actuator, and the deviation range exceeds the set threshold value of the opening feedback range (for example, 5%), then the electronic device determines that the actuator has a stuck fault. Thus, the embodiment of the present invention realizes the fault analysis of the actuator. It provides an effective basis for further control optimization, efficiency improvement, energy conservation and emission reduction.

[0168] Please note that, please refer to Figure 6 , the embodiment of the present invention can separate the control performance evaluation from the fault diagnosis, and divide the evaluation process into two cycles, namely the control performance evaluation cycle and the fault diagnosis cycle; the control performance evaluation cycle does not require high data real-time performance, and a large amount of historical data is required in the index calculation process, so the cycle is usually about 30 minutes to 60 minutes, or even longer; while the fault diagnosis cycle has a high requirement for real-time performance and should reflect the fault status as soon as possible, usually 30 seconds to 60 seconds, to ensure that the diagnosis result is obtained quickly.

[0169] In summary, the embodiment of the present invention achieves comprehensive coverage of the control performance evaluation of thermal power units, and can integrate the performance evaluation that was previously focused on a certain link, and finally conduct a comprehensive control performance evaluation of each part of the entire thermal power unit. This provides an effective basis for further control optimization, efficiency improvement, and energy conservation and emission reduction. At the same time, with the storage of massive data, it also lays a good foundation for the application of data mining and big data technology in the field of thermal control of thermal power plants.

[0170] Device Embodiment

[0171] Please refer to Figure 7 The present invention provides a thermal power unit control performance evaluation device, comprising:

[0172] The data acquisition module 701 is used to acquire the historical data of the distributed control system of the thermal power unit within a preset time period;

[0173] A state identification module 702 is used to identify the state of the distributed control system historical data based on the data change trend to obtain a dynamic adjustment process data set and a steady-state random process data set;

[0174] A performance evaluation module 703, configured to calculate a deterministic performance evaluation index based on the dynamic adjustment process data set, and to calculate a random performance evaluation index based on the steady-state random process data set;

[0175] Among them, the data change trend of the dynamic adjustment process data set is greater than the data change trend of the steady-state random process data set; the deterministic performance evaluation index characterizes the dynamic adjustment performance of the thermal power unit, and the random performance evaluation index characterizes the performance of the thermal power unit in resisting random disturbances in the steady-state process.

[0176] Compared with the existing scheme that directly uses the adjustment time, rise time and overshoot and other indicators to measure the control performance of the thermal power unit, it is easily affected by large disturbances, resulting in reduced accuracy of control performance evaluation. The present invention divides the historical data of the distributed control system into a dynamic adjustment process data set in a dynamic adjustment process and a steady-state random process data set in a steady-state random process according to the data change trend, and then calculates the deterministic performance evaluation index in the dynamic adjustment process and the random performance evaluation index in the steady-state random process. By dividing the historical data of the distributed control system into two parts of data to calculate the control performance, the influence of large disturbances is reduced, thereby improving the accuracy of the control performance evaluation of the thermal power unit.

[0177] Optionally, the state identification module 702 includes:

[0178] A piecewise linearization processing module, used for performing piecewise linearization processing on the historical data of the distributed control system to obtain piecewise linearization data;

[0179] The identification module is used to perform state identification on the piecewise linearization data based on the data change trend of each straight line segment in the piecewise linearization data to obtain a dynamic adjustment process data set and a steady-state random process data set.

[0180] Optionally, the identification module includes:

[0181] A straight line slope calculation module, used to calculate the straight line slope of each straight line segment in the piecewise linearization data;

[0182] A straight line slope change calculation module, used to determine the straight line slope change of adjacent straight lines in the piecewise linearization data based on the straight line slopes of the straight line segments;

[0183] The final identification module is used to identify the change in the slope of the straight line based on a set threshold value to obtain a dynamic adjustment process data set and a steady-state random process data set.

[0184] Optionally, the performance evaluation module 703 includes:

[0185] An autoregressive model building module is used to perform time series analysis based on the steady-state random process data set and build an autoregressive model;

[0186] A theoretical minimum variance calculation module, used to determine the theoretical minimum variance of a steady-state random process data set based on the autoregressive model;

[0187] The first performance index calculation module is used to calculate the minimum variance based on the theoretical minimum variance and the actual output variance of the steady-state random process data set to obtain the randomness performance evaluation index.

[0188] Optionally, the performance evaluation module 703 includes:

[0189] A linear prediction model building module, used to build a linear prediction model based on the dynamic adjustment process data set;

[0190] An estimated value calculation module, used for estimating model parameters of the linear prediction model based on a gradient descent method to obtain an estimated value;

[0191] A residual check module, used for performing residual check on the linear prediction model based on the estimated value;

[0192] The second performance indicator calculation module is used to calculate the rise time, the adjustment time and the overshoot based on the linear prediction model after residual verification, so as to serve as the deterministic performance evaluation indicators.

[0193] Optionally, the distributed control system historical data includes sensor historical data; and the thermal power unit control performance evaluation device further includes:

[0194] The first fault analysis module is used to perform fault analysis on the rate of change of the value of the historical data of the sensor based on a set change rate threshold.

[0195] Optionally, the distributed control system historical data includes historical opening instructions of the actuator; the thermal power unit control performance evaluation device also includes:

[0196] The second fault analysis module is used to perform fault analysis on the actuator based on the historical opening instructions and the opening feedback corresponding to the historical opening instructions.

[0197] The thermal power unit control performance evaluation device includes a processor and a memory. The data acquisition module 701, the state identification module 702 and the performance evaluation module 703 are all stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.

[0198] The processor includes a kernel, which calls the corresponding program unit from the memory. There can be one or more kernels.

[0199] 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.

[0200] Figure 8 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 8As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 complete mutual communication through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the control performance evaluation method of the thermal power unit, the method comprising: obtaining the distributed control system historical data of the thermal power unit within a preset time period; performing state recognition on the distributed control system historical data based on the data change trend to obtain a dynamic adjustment process data set and a steady-state random process data set; calculating a deterministic performance evaluation index based on the dynamic adjustment process data set, and calculating a random performance evaluation index based on the steady-state random process data set; wherein the data change trend of the dynamic adjustment process data set is greater than the data change trend of the steady-state random process data set; the deterministic performance evaluation index represents the dynamic adjustment performance of the thermal power unit, and the random performance evaluation index represents the performance of the thermal power unit against random disturbances in the steady-state process.

[0201] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0202] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a machine-readable storage medium. When the computer program is executed by a processor, the computer can execute the thermal power unit control performance evaluation method provided by the above-mentioned methods, and the method includes: obtaining the historical data of the distributed control system of the thermal power unit within a preset time period; performing state identification on the historical data of the distributed control system based on the data change trend to obtain a dynamic adjustment process data set and a steady-state random process data set; calculating a deterministic performance evaluation index based on the dynamic adjustment process data set, and calculating a random performance evaluation index based on the steady-state random process data set; wherein the data change trend of the dynamic adjustment process data set is greater than the data change trend of the steady-state random process data set; the deterministic performance evaluation index characterizes the dynamic adjustment performance of the thermal power unit, and the random performance evaluation index characterizes the performance of the thermal power unit against random disturbances in the steady-state process.

[0203] On the other hand, the present invention also provides a machine-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the control performance evaluation method of a thermal power unit provided by the above-mentioned methods, the method comprising: obtaining historical data of a distributed control system of the thermal power unit within a preset time period; performing state identification on the historical data of the distributed control system based on a data change trend to obtain a dynamic adjustment process data set and a steady-state random process data set; calculating a deterministic performance evaluation index based on the dynamic adjustment process data set, and calculating a random performance evaluation index based on the steady-state random process data set; wherein the data change trend of the dynamic adjustment process data set is greater than the data change trend of the steady-state random process data set; the deterministic performance evaluation index characterizes the dynamic adjustment performance of the thermal power unit, and the random performance evaluation index characterizes the performance of the thermal power unit against random disturbances in a steady-state process.

[0204] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0205] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0206] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating control performance of a thermal power unit, characterized in that: The method comprises: Obtaining historical data of the distributed control system of the thermal power unit within a preset time period; Based on the data change trend, the state of the historical data of the distributed control system is identified to obtain a dynamic adjustment process data set and a steady-state random process data set; Calculating a deterministic performance evaluation index based on the dynamic adjustment process data set, and calculating a stochastic performance evaluation index based on the steady-state random process data set; Among them, the data change trend of the dynamic adjustment process data set is greater than the data change trend of the steady-state random process data set; the deterministic performance evaluation index characterizes the dynamic adjustment performance of the thermal power unit, and the random performance evaluation index characterizes the performance of the thermal power unit in resisting random disturbances in the steady-state process.

2. The control performance evaluation method of a thermal power unit according to claim 1, characterized in that: The state identification of the historical data of the distributed control system based on the data change trend is performed to obtain a dynamic adjustment process data set and a steady-state random process data set, including: Performing piecewise linearization processing on the historical data of the distributed control system to obtain piecewise linearization data; Based on the data variation trend of each straight line segment in the piecewise linearization data, the state of the piecewise linearization data is identified to obtain a dynamic adjustment process data set and a steady-state random process data set.

3. The control performance evaluation method of a thermal power unit according to claim 2, characterized in that: The state recognition is performed on the piecewise linearization data based on the data change trend of each straight line segment in the piecewise linearization data to obtain a dynamic adjustment process data set and a steady-state random process data set, including: Calculating the slope of each straight line in the piecewise linearization data; Determine the change in the slope of each straight line of the piecewise linearization data based on the slope of each straight line; The change in the slope of the straight line is identified based on a set threshold, and a dynamic adjustment process data set and a steady-state random process data set are obtained.

4. The method for evaluating control performance of a thermal power unit according to claim 1, characterized in that: The calculating of the randomness performance evaluation index based on the steady-state random process data set includes: Performing time series analysis based on the steady-state random process data set and establishing an autoregressive model; Determining the theoretical minimum variance of a steady-state random process data set based on the autoregressive model; Based on the theoretical minimum variance and the actual output variance of the steady-state random process data set, the minimum variance is calculated to serve as the randomness performance evaluation index.

5. The method for evaluating control performance of a thermal power unit according to claim 1, characterized in that: The calculating of a deterministic performance evaluation index based on the dynamic adjustment process data set includes: Constructing a linear prediction model based on the dynamic adjustment process data set; Estimating model parameters of the linear prediction model based on a gradient descent method to obtain estimated values; Performing residual check on the linear prediction model based on the estimated value; Based on the residual-checked linear prediction model, the rise time, the adjustment time and the overshoot are calculated to serve as the deterministic performance evaluation indicators.

6. The method for evaluating control performance of a thermal power unit according to claim 1, characterized in that: The distributed control system historical data includes sensor historical data; after acquiring the distributed control system historical data of the thermal power unit within a preset time period, the method further includes: A fault analysis is performed on the rate of change of the value of the historical data of the sensor based on a set rate of change threshold.

7. The method for evaluating control performance of a thermal power unit according to claim 1, characterized in that: The distributed control system historical data includes the historical opening instructions of the actuator; after the distributed control system historical data of the thermal power unit within a preset time period is obtained, it also includes: A fault analysis is performed on the actuator based on the historical opening instructions and the opening feedback corresponding to the historical opening instructions.

8. A thermal power unit control performance evaluation device, characterized in that: include: A data acquisition module is used to acquire the historical data of the distributed control system of the thermal power unit within a preset time period; A state identification module, used to identify the state of the historical data of the distributed control system based on the data change trend, and obtain a dynamic adjustment process data set and a steady-state random process data set; A performance evaluation module, used to calculate a deterministic performance evaluation index based on the dynamic adjustment process data set, and to calculate a stochastic performance evaluation index based on the steady-state random process data set; Wherein, the data variation trend of the dynamic adjustment process data set is greater than the data variation trend of the steady-state random process data set; The deterministic performance evaluation index characterizes the dynamic regulation performance of the thermal power unit, and the stochastic performance evaluation index characterizes the performance of the thermal power unit in resisting random disturbances in a steady-state process.

9. 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 program, the method for evaluating the control performance of a thermal power unit as described in any one of claims 1 to 7 is implemented.

10. A machine-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for evaluating the control performance of a thermal power unit according to any one of claims 1 to 7 is implemented.

Images