An air compressor surge prevention auto-disturbance control method

Abstract

The present invention belongs to the field of information technology, and proposes an air compressor surge prevention self-disturbance control method. The present invention starts with dimensionless anti-surge control, firstly calculates and draws the characteristic curve of the air compressor for the actual operation data of the actual air compressor, and converts it into a simplified flow and pressure head characteristic, and then calculates the surge boundary line and the anti-surge control line according to the simplified flow and pressure head characteristics of the air compressor, and further calculates the real-time surge proximity and surge deviation, and finally designs the corresponding self-disturbance control controller according to the surge deviation to perform anti-surge control on the air compressor. The present invention does not rely on the precise model of the object, does not need to consider dynamic coupling, is easy to implement, has high precision and strong anti-disturbance capability; effectively prevents the generation of air compressor surge, and ensures the rapidity and robustness of control; this method can be widely used in different industrial fields.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to an air compressor surge prevention self-disturbance control method. Background Art

[0002] Surge means that the gas pressure in the output pipeline is higher than the compressor outlet pressure. The compressed gas will flow back into the compressor, and its outlet pressure and suction flow will produce low-frequency, periodic, and large-scale fluctuations, which can cause serious damage to the compressor. It needs to be prevented and controlled to prevent surge. At present, the most commonly used method for anti-surge control of air compressors is PID control. This method can effectively prevent the occurrence of surge and has achieved significant economic benefits. However, the PID control parameters are complicated and cumbersome, and it is difficult to achieve satisfactory results in places with high control speed requirements. Therefore, a control method is urgently needed in the actual industrial field to make up for the shortcomings of PID control.

[0003] In order to avoid surge, the solution of adjusting the opening of the return valve is usually adopted to meet the actual operating state of the compressor and the needs of the compressor operating conditions. (Gravdahl JT, Egeland O, Vatland SO. Drive torque actuation in active surge control of centrifugal compressors [J]. Pergamon Press, Inc. 2002 (11).) A PID anti-surge control method based on the mass flow rate of the air compressor and its axial speed was designed; (Chen Huaizhong, Jiang Lei. Research on adaptive genetic PID algorithm in compressor anti-surge [J]. Control Engineering, 17 (6): 3.) Based on genetic algorithm, an adaptive PID control algorithm was designed to control compressor surge; (G. Torrisi, S. Grammatico, A. Cortinovis, M. M.Morari and R.S. Smith, "Model Predictive Approaches for Active Surge Control in Centrifugal Compressors," in IEEE Transactions on Control Systems Technology, vol.25, no.6, pp.1947-1960, Nov.2017,) designed a centrifugal compressor anti-surge control method based on model predictive control; (X.Ma, S.Zheng and K.Wang, "Active Surge Control for Magnetically Suspended Centrifugal Compressors Using a Variable Equilibrium Point Approach," in IEEE Transactions on Industrial Electronics, vol.66, no.12, pp.9383-9393, Dec.2019) established a robust simplified system of the magnetic bearing control system, and designed the corresponding robust controller using the μ synthesis method.

[0004] The anti-surge control of air compressors requires fast and stable operation. The research on anti-surge controllers for air compressors generally follows the developed control algorithms. The most commonly used PID controller in factories has achieved certain economic benefits, but it is very difficult to tune the PID anti-surge control in the presence of external interference. For some industrial scenarios with high control speed requirements, the PID controller is relatively difficult to control. At present, there is an urgent need for an anti-surge control method that can solve the above problems at the same time. Summary of the invention

[0005] The main problem solved by the present invention is the surge prevention and control problem during the operation of the air compressor. In order to solve the above problem, an actual industrial air compressor system is analyzed, and the air compressor characteristic curve is drawn according to the real-time operation database of the air compressor, and the surge limit line (SLL) and surge control line (SCL) of the air compressor are determined. The surge proximity of the air compressor is further determined using the real-time output flow and compression ratio; the surge deviation is calculated using the surge proximity and surge margin, and the surge deviation is controlled using the anti-disturbance controller to achieve anti-surge control of the air compressor.

[0006] The technical solution of the present invention is as follows: an air compressor surge prevention auto-disturbance rejection control method, the specific steps are as follows:

[0007] Step 1: Read the flow rate and output pressure data from the air compressor operation database, and calculate the real-time compression ratio and flow rate of the air compressor;

[0008] Step 2: Calculate the air compressor characteristic curve;

[0009] According to the polytropic pressure head formula and the inlet flow square formula:

[0010]

[0011]

[0012]

[0013] Among them, Q s is the flow rate, H p is the polytropic pressure head, ε is the compression ratio, Z avg =(Z s +Z d ) / 2,Z s is the compressor inlet compression factor, Z d is the compressor outlet compression factor, T s is the compressor inlet temperature, R is the compressor outlet compression factor, T s is the compressor inlet temperature, T d is the compressor outlet temperature. According to experience, Z avg With Z s The ratio change of can be ignored, assuming that is equal within the compressor operating range, the simplified pressure head and simplified flow square formula can be obtained:

[0014]

[0015]

[0016] Among them, q r represents the simplified flow, h r To simplify the pressure head; dP os is the compressor inlet flow orifice differential pressure, P s is the compressor inlet pressure; using the above formula, the horizontal axis is the simplified flow rate squared q r 2 , the vertical axis is the simplified pressure head h r of Air compressor characteristic curve;

[0017] Step 3: Calculate the surge limit line SLL and the anti-surge control line SCL;

[0018] According to the air compressor characteristic curve in step 2, the surge limit line SLL is calculated, which is the line formed by the vertices of the characteristic curve of the air compressor at different speeds, which is approximately equal to the surge boundary line; in order to prevent surge, control must be applied before the operating point reaches the surge limit line SLL, so a surge control margin SCM is set outside the surge limit line SLL; an initial anti-surge control line SCL is set on the right side of the surge limit line SLL, and the margin between the initial anti-surge control line SCL and the surge limit line SLL is SCM;

[0019] SCM=-(b+nb0)f4(Z) (6)

[0020] In the formula, b is the initial size of the safety margin, parameter n is the number of surges, b0 is the margin that needs to be increased each time a surge occurs, and f4(Z) is a linear characteristic function, which is selected according to the characteristics of the air compressor and the actual operating characteristics of the air compressor and can be taken as 1;

[0021] Step 4: Calculate the surge proximity S s ;

[0022] Calculate surge proximity based on simplified flow and compression ratio:

[0023]

[0024] In the formula, K is a constant, which is selected according to the actual characteristics of the air compressor;

[0025] When S s =1, the operating point is exactly on the surge line, and the air compressor surges;

[0026] When S s When >1, the actual flow rate is less than the surge flow rate, and the air compressor will surge;

[0027] When S s When <1, the actual flow rate is greater than the surge flow rate, and the air compressor will not surge;

[0028] Step 5: Calculate the number of surges;

[0029] According to the air compressor characteristic curve in step 2 and the surge limit line SLL in step 3, calculate the distance d of the current time value from the air compressor characteristic curve SLL in the air compressor characteristic curve, and calculate the equivalent speed v at this time:

[0030]

[0031] In the formula, q r0 is the simplified flow at the current moment, h r0 Simplify the pressure head at the current moment, t represents time; determine the equivalent distance l that the operating point at the current moment can move at the current equivalent speed:

[0032] l=v×Δt (9)

[0033] Δt represents the sampling time;

[0034] When the equivalent distance l is greater than the surge control margin SCM, n=n+1, and SCM is updated;

[0035] Step 5: Calculate the deviation value DEV;

[0036] The air compressor deviation DEV is introduced as the control quantity, that is, the distance between the operating point and the initial anti-surge control line SCL:

[0037] DEV=1-SCM-S s (10)

[0038] When DEV=0, the distance between the operating point and the surge limit is equal to SCM; when DEV>0, the distance between the operating point and the surge limit is greater than SCM, and anti-surge control is not performed at this time; when DEV<0, the distance between the operating point and the surge limit is less than SCM, and anti-surge control is performed on the air compressor;

[0039] Step 6: Design an active disturbance rejection controller, that is, establish the system linear extended state observer LESO and control rate design;

[0040] According to the model characteristics of the air compressor, the following linear extended state observer is established:

[0041]

[0042] in: y represents the actual output of the controlled object, which is taken as DEV, b 10 is the controller gain, represents the estimated output of the linear extended state observer, z represents the state estimate, L represents the observation matrix, and C is the system output matrix. According to the design principle of the full-dimensional state observer, the poles of the characteristic equation in equation (10) are configured to satisfy the error attenuation rate, and its poles are configured at w0, resulting in:

[0043] |sI-(A-LC)|=s 3 +l1s 2 +l2s+l3=(s+w0) 3 (11)

[0044]

[0045] Where, s is the complex frequency, w0 is the bandwidth of the linear extended state observer, and is the setting parameter;

[0046] The control rate is taken as:

[0047]

[0048] where u0 is given by the PD combination, k p , represents the controller parameter to be tuned, which is k d1 =2w c , k d2 =1, ref represents the set value DEV0; w c Represents the controller bandwidth, z1, z2, z3 represent the state output of the expanded state observer, and repeat steps 4 to 6 to perform surge prevention control on the air compressor.

[0049] The beneficial effects of the present invention are as follows: when the present invention controls the actual air compressor in the industrial field, the flow rate and the compression ratio are converted to obtain the corresponding dimensionless data, and the air compressor is controlled against surge by using the self-disturbance rejection controller, which can respond to the real-time status of the air compressor faster and more quickly, and the control effect is significantly improved; and the control by using the self-disturbance rejection controller does not rely on the precise model of the object, does not need to consider dynamic coupling, is easy to implement, has high accuracy and strong anti-disturbance ability. According to the characteristic curve of the air compressor, the self-disturbance rejection anti-surge control based on the characteristics of the air compressor is designed, which solves the problem of quickly realizing anti-surge control, prevents the harm caused by surge phenomenon to industrial practice, and provides technical support for the anti-surge control of air compressors in industrial fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 The present invention is a flowchart of an implementation of an air compressor surge prevention self-disturbance control method.

[0051] Figure 2(a) is a characteristic curve of the air compressor.

[0052] Figure 2(b) is the anti-surge control curve.

[0053] Figure 3(a) is the anti-surge control flow curve.

[0054] Figure 3(b) is the anti-surge control compression ratio curve. DETAILED DESCRIPTION

[0055] The related theories such as self-disturbance control, air compressor surge modeling driven by mechanism and data collaboration are an air compressor surge prevention self-disturbance control method. The present invention starts with the air compressor surge generation mechanism and historical data. First, according to the air compressor characteristic curve, the actual compression ratio and flow data are converted into simplified pressure head and simplified flow data, and its simplified flow and pressure head characteristics are further determined. The surge boundary line and anti-surge control line are calculated, and the real-time surge proximity and surge deviation are calculated. Finally, according to the surge deviation, the corresponding self-disturbance control controller is designed to perform anti-surge control on the air compressor. The air compressor surge prevention self-disturbance control method can effectively prevent the generation of air compressor surge, and improve the rapidity and robustness of the unit control. It can be promoted and applied to air compressor units of different specifications for anti-surge control.

[0056] The overall implementation process of the technical solution of the present invention is shown in the attached Figure 1 As shown, the specific steps are as follows:

[0057] Data processing: extract flow and output pressure data from the industrial site real-time database, calculate the output compression ratio, and convert the data into simplified pressure head and flow data;

[0058] Characteristic curve drawing: Integrate the simplified pressure head and flow data obtained by processing, draw the characteristic curve of the air compressor, and calculate the surge limit line SLL and the initial anti-surge control line SCL;

[0059] Calculate the real-time surge proximity S based on the real-time flow and compression ratio s ;

[0060] Given the initial surge margin SCM0, according to the previously obtained air compressor characteristic curve and surge limit line SLL, calculate the distance d of the moment value from SLL in the characteristic curve, and calculate the equivalent speed v at this time, and determine whether the margin requirement is met. If not, update the surge margin SCM;

[0061] According to the obtained surge proximity S s Calculate the surge deviation DEV with the surge margin SCM;

[0062] According to the DEV obtained previously, a linear active disturbance rejection controller is designed to perform anti-surge control on the air compressor.

[0063] Taking the compressor fan stall model Moore-Greitzer (MG) as an example, the implementation mode of the present invention is described in detail in combination with the accompanying drawings. By analyzing the low-order MG model, the model is integrated and improved, the anti-surge valve opening variable and the air compressor shaft speed variable are added, and compared with the actual compressor parameters of a certain factory, the characteristic curve under this group of parameters is calculated, and its surge and stall characteristics are calculated. Based on the model, a corresponding anti-surge controller is designed. In order to better understand the technical solution of the present invention, the specific implementation steps of the present invention are as follows:

[0064] Step 1: Data processing;

[0065] The MG model was tested several times to obtain its flow rate and compression ratio at different speeds, and the data were converted into simplified pressure head and flow rate data;

[0066] Step 2: Calculate the air compressor characteristic curve;

[0067] According to the calculated compression ratio and flow rate, the following conversion calculation is performed:

[0068]

[0069]

[0070]

[0071] Among them, q r represents the simplified flow, Q s is the actual flow rate, h r To simplify the pressure head, ε is the compression ratio, Z s is the compressor inlet compression factor, T s is the compressor inlet temperature, T d is the compressor outlet temperature, R is the compressor outlet compression factor, dP os is the compressor inlet flow orifice differential pressure, P s is the compressor inlet pressure. The horizontal axis is the simplified flow rate squared q r 2 , the vertical axis is the simplified pressure head h r curve and determine the surge boundary line;

[0072] Step 3: Calculate the surge proximity S s ;

[0073] Calculate the real-time surge proximity S based on the real-time flow and compression ratio s , calculated as follows:

[0074]

[0075] Select K according to the actual characteristics of the air compressor and set the initial margin SCM0.

[0076] Step 4: Calculate the surge margin SCM;

[0077] According to the compressor characteristic curve and surge limit line SLL obtained in step 3, calculate the distance d from SLL in the characteristic curve, and calculate the equivalent speed v at this time. The calculation method is as follows:

[0078]

[0079] In the formula, q r0 is the simplified flow at the current moment, h r0 is the simplified pressure head at the current moment, and t represents time. Further determine the equivalent distance l that the operating point at the current moment can move at the current equivalent speed:

[0080] l=v×Δt (6)

[0081] Δt represents the sampling time. If at this time: l>SCM, then n=n+1, and the SCM is updated and calculated.

[0082] SCM=-(b+nb0)f4(Z) (7)

[0083] Wherein, the initial value of the safety margin b is SCM0, n is the number of surge occurrences, b0 is the margin that needs to be increased each time surge occurs, and f4(Z) is a linear characteristic function, which can be taken as 1 according to the actual air compressor characteristics.

[0084] Step 5: Calculate the deviation value DEV;

[0085] Introduce the air compressor deviation DEV, that is, the distance between the operating point and SCL:

[0086] DEV=1-SCM-S s (8)

[0087] Step 6: Design an ADRC controller;

[0088] According to the characteristics of the compressor fan stall model, the following linear extended state observer is established:

[0089]

[0090] in, y represents the actual output of the controlled object, which is taken as DEV in the present invention. Represents the estimated output, z represents the state estimate, L represents the observation matrix, and C is the system output matrix. Configure the poles of the characteristic equation to satisfy the error attenuation rate, and configure its poles at w0, we can get:

[0091] |sI-(A-LC)|=s 3 +l1s 2 +l2s+l3=(s+w0) 3 (10)

[0092]

[0093] Where, s is the complex frequency, w0 is the bandwidth of the linear extended state observer, and is the setting parameter;

[0094] The control rate is taken as:

[0095]

[0096] Where u0 is given by the PD combination u0 = k p (r-z1)-k d1 z2-k d2 z3, k p , represents the controller parameters to be tuned, z1, z2, z3 represent the state outputs of the extended state observer, and are taken as k d1 =2w c , k d2 =1, r represents the set value DEV0; repeat steps 3 to 6 to perform surge prevention control on the air compressor. Table 1 shows the comparison between the parameter adjustment results of the present invention and other methods:

[0097] Table 1 Comparison of anti-surge effects between the method of the present invention and other methods

[0098]

[0099] The invention can be used to implement surge prevention and control of air compressors, effectively solve the problem of rapid implementation of surge control, effectively avoid the occurrence of surge, prevent the harm caused by surge to industrial reality, and provide technical support for air compressor anti-surge control work at industrial sites.

Claims

1. An air compressor surge prevention auto-disturbance rejection control method, characterized in that: The specific steps are as follows: Step 1: Read the flow rate and output pressure data from the air compressor operation database, and calculate the real-time compression ratio and flow rate of the air compressor; Step 2: Calculate the air compressor characteristic curve; According to the polytropic pressure head formula and the inlet flow square formula: Among them, Q s is the flow rate, H p is the polytropic pressure head, ε is the compression ratio, Z avg =(Z s +Z d ) / 2,Z s is the compressor inlet compression factor, Z d is the compressor outlet compression factor, T s is the compressor inlet temperature, R is the compressor outlet compression factor, T s is the compressor inlet temperature, T d is the compressor outlet temperature; Z avg With Z s The ratio is equal within the compressor operating range, resulting in a simplified head and simplified flow square formula: Among them, q r represents the simplified flow, h r To simplify the pressure head; dP os is the compressor inlet flow orifice differential pressure, P s is the compressor inlet pressure; using the above formula, the horizontal axis is the simplified flow rate square q r 2 , the vertical axis is the simplified pressure head h r of Air compressor characteristic curve; Step 3: Calculate the surge limit line SLL and the anti-surge control line SCL; According to the air compressor characteristic curve in step 2, the surge limit line SLL is calculated, which is the line formed by the vertices of the characteristic curve of the air compressor at different speeds; in order to prevent surge, control must be applied before the operating point reaches the surge limit line SLL, so a surge control margin SCM is set outside the surge limit line SLL; an initial anti-surge control line SCL is set on the right side of the surge limit line SLL, and the margin between the initial anti-surge control line SCL and the surge limit line SLL is SCM; SCM=-(b+nb0)f4(Z)(6) Where, b is the initial size of the safety margin, parameter n is the number of surges, b0 is the margin that needs to be increased each time a surge occurs, and f4(Z) is a linear characteristic function, which is selected based on the characteristics of the air compressor and the actual operating characteristics of the air compressor; Step 4: Calculate the surge proximity S s ; Calculate surge proximity based on simplified flow and compression ratio: In the formula, K is a constant, which is selected according to the actual characteristics of the air compressor; When S s =1, the operating point is exactly on the surge line, and the air compressor surges; When S s When >1, the actual flow rate is less than the surge flow rate, and the air compressor will surge; When S s When <1, the actual flow rate is greater than the surge flow rate, and the air compressor will not surge; Step 5: Calculate the number of surges; According to the air compressor characteristic curve in step 2 and the surge limit line SLL in step 3, calculate the distance d of the current time value from the air compressor characteristic curve SLL in the air compressor characteristic curve, and calculate the equivalent speed v at this time: In the formula, q r0 is the simplified flow at the current moment, h r0 Simplify the pressure head at the current moment, t represents time; determine the equivalent distance l that the operating point at the current moment can move at the current equivalent speed: l=v×Δt(9) Δt represents the sampling time; When the equivalent distance l is greater than the surge control margin SCM, n=n+1, and SCM is updated; Step 5: Calculate the deviation value DEV; The air compressor deviation DEV is introduced as the control quantity, that is, the distance between the operating point and the initial anti-surge control line SCL: DEV=1-SCM-S s (10) When DEV=0, the distance between the operating point and the surge limit is equal to SCM; when DEV>0, the distance between the operating point and the surge limit is greater than SCM, and anti-surge control is not performed at this time; when DEV<0, the distance between the operating point and the surge limit is less than SCM, and anti-surge control is performed on the air compressor; Step 6: Design an active disturbance rejection controller, that is, establish the system linear extended state observer LESO and control rate design; According to the model characteristics of the air compressor, the following linear extended state observer is established: in: y represents the actual output of the controlled object, which is taken as DEV, b 10 is the controller gain, represents the estimated output of the linear extended state observer, z represents the state estimate, L represents the observation matrix, and C is the system output matrix. According to the design principle of the full-dimensional state observer, the poles of the characteristic equation in equation (10) are configured to satisfy the error attenuation rate, and its poles are configured at w0, resulting in: sI-(A-LC)=s 3 +l1s 2 +l2s+l3=(s+w0) 3 (11) Where, s is the complex frequency, w0 is the bandwidth of the linear extended state observer, and is the setting parameter; The control rate is taken as: Where u0 is given by the PD combination, u0 = k p (ref-z1)-k d1 z2-k d2 z3, k p ,k d1 ,k d2 represents the controller parameter to be tuned, which is k d1 =2w c , k d2 =1, ref represents the set value DEV0; w c Represents the controller bandwidth, z1, z2, z3 represent the state output of the expanded state observer, and repeat steps 4 to 6 to perform surge prevention control on the air compressor.

Images