CN105736059A - High-speed dynamic balance ability optimization design method for gas turbine pull rod rotor with end face teeth - Google Patents

Abstract

The invention discloses a high-speed dynamic balance ability optimization design method for a gas turbine pull rod rotor with end face teeth. The high-speed dynamic balance ability optimization design method is used for optimizing the high-speed dynamic balance ability of the gas turbine pull rod rotor. According to the method, an objective function for optimizing the installation angles of wheel discs of the gas turbine rotor is provided, the installation angles of the wheel discs are optimized through the genetic algorithm, the unbalance centrifugal force on the gas turbine rotor and the bending moment exerted on the midpoint of the rotor by the unbalance centrifugal force are lowered, the multi-objective genetic algorithm is adopted for optimizing the installation angles of the wheel discs of the gas turbine rotor, the unbalance response amplitude of the rotor is obviously lowered after optimization than that before optimization, and the axle vibration amplitude on a bearing of the rotor is far smaller than the allowable maximum amplitude of dynamic balance. Theoretically, the rotor can directly pass in-plant high-speed dynamic balance without other means such as balance weights through the optimization scheme, and the in-plant high-speed dynamic balance efficiency is improved. According to the method, the in-plant dynamic balance testing ability and efficiency of the rotor can be obviously improved for manufacturing enterprises.

Description

Gas turbine pull rod rotor high-speed balancing ability Optimization Design with end-tooth

Technical field:

The present invention relates to gas turbine technology field, particularly to the gas turbine pull rod rotor high-speed balancing ability Optimization Design with end-tooth.

Background technology:

Heavy duty gas turbine rod fastening rotor is a kind of typical composite fabricated rotor, by a center pull rod or many circumferential pull bar traverse wheel discs at different levels, by pull bar being applied pretightning force and the stay-bolt of tight two ends spindle nose, wheel disc is compressed to be combined as a whole by combined rotor.Owing to the rotor weight of this structure is light, it is easily assembled and there is good cooling effect, being widely applied in gas turbine engine and aero-engine.Machining inaccuracy in gas turbine wheel disk manufacture process so that there is amount of unbalance on rotor, at high speed, produces very big uneven exciting force, causes unit vibration.For the gas turbine rotor structure that end-tooth connects, by adjusting end-tooth setting angle, it is possible to improve the efficiency of rotor high-speed balancing.

Summary of the invention:

It is an object of the invention to provide a kind of gas turbine pull rod rotor high-speed balancing ability Optimization Design with end-tooth, the method provides the relation between wheel disc radial run-out and rotor unbalance value, and the amount of unbalance for rotor determines offer reference.And the relation according to both, optimizing wheel disc setting angle at different levels, the uneven centrifugal force and the unbalanced moments that produce amount of unbalance are minimized, and then reduce rotor bearing place vibratory response amplitude, reduce the purpose of rotor oscillation.

For reaching above-mentioned purpose, the present invention adopts the following technical scheme that and realizes:

With the gas turbine pull rod rotor high-speed balancing ability Optimization Design of end-tooth, comprise the following steps:

1) the radial run-out value e according to the front spindle nose of gas turbine pull rod rotor, rear spindle nose and wheel disc at different levels, it is determined that the initial size of whole rotor difference parts amount of unbalance and PHASE DISTRIBUTION；

2) according to the initial size of amount of unbalance and PHASE DISTRIBUTION, it is determined that uneven centrifugal force to be optimized and each uneven centrifugal force are object function to the moment of flexure of rotor midpoint；

3) work out genetic algorithm optimization program according to object function, adopt genetic algorithm to obtain the minima of object function and the wheel disc end-tooth setting angle of correspondence thereof；

4) by contrasting initial settling angle degree and optimizing the unbalance response of setting angle lower rotor part, after optimization, axle amplitude first-order kernel peak-fall amplitude in compressor end bearing place is more than 95%, and turbine stub shaft holds place's axle amplitude first-order kernel peak-fall amplitude more than 95%；After optimization, axle amplitude second-order response peak-fall amplitude in compressor end bearing place is more than 95%, and turbine stub shaft holds place's axle amplitude second-order response peak-fall amplitude more than 80%, it is determined that the genetic algorithm effectiveness to gas turbine rotor high-speed balancing.

The present invention is further improved by, and described gas turbine pull rod rotor, with end face tooth structure, is connected by end-tooth between adjacent wheel disc, and end-tooth attachment structure is used for adjusting wheel disc setting angle.

The present invention is further improved by, step 1) in, the relation between radial run-out degree e and amount of unbalance is:

Wherein, q is amount of unbalance, and e is radial run-out degree, and m is this grade of wheel disc quality.

The present invention is further improved by, step 2) in, optimization object function is:

The vector of uneven centrifugal force and out-of-balance force moment of flexure is expressed as:

Wherein, q (i) represents the amount of unbalance/g mm of wheel disc at different levels；α (i) represent amount of unbalance phase place/°, for the variable in genetic algorithm, namely corresponding wheel disc setting angle；K represents wheel disc progression；ω is rotor speed/r min^-1；L (i) represents the wheel disc amount of unbalance at different levels distance/mm to rotor midpoint position；I span is 1～25；F_xFor x direction imbalance centrifugal force/N；F_yFor y direction imbalance centrifugal force/N；F is total uneven centrifugal force/N；M_xFor x direction out-of-balance force moment of flexure/N m；M_yFor y direction out-of-balance force moment of flexure/N m；M is total out-of-balance force moment of flexure/N m.

Compared with prior art, the beneficial effects of the present invention is:

The present invention gives the gas turbine pull rod rotor dynamic balancing ability Optimization Design with end face tooth structure, the rotor dynamic balancing ability optimization with end face tooth structure is provided a kind of method by adjusting wheel disc end-tooth setting angle by the method.By this optimization method, before and after optimizing, compressor end bearing place axle amplitude first-order kernel peak-fall amplitude is more than 95%, and turbine stub shaft holds place's axle amplitude first-order kernel peak-fall amplitude more than 95%；Before and after optimizing, compressor end bearing place axle amplitude second-order response peak-fall amplitude is more than 95%, and turbine stub shaft holds place's axle amplitude second-order response peak-fall amplitude more than 80%, and therefore the method can be effectively improved manufacturer and carries out the ability of rotor dynamic balancing.The method is applicable to the relevant industries such as space flight, electric power with the rotor part of end face tooth structure, has future in engineering applications widely.

Accompanying drawing illustrates:

Fig. 1 is certain center pull rod gas turbine rotor typical structure schematic diagram, gives rotor end-face tooth and center pull rod structure in figure.

Fig. 2 is gas turbine wheel disk amount of unbalance schematic diagram.

Fig. 3 is gas turbine rotor each position circular runout schematic diagram.

Providing 25 circular runout positions in figure altogether, wherein S1, S2, S3 are spindle nose position, front and back radius run-out degree, C1, C2 ... C15 is compressor disk radius run-out degree, N1, N2, and N3 is jackshaft radius run-out degree, T1, T2, T3, and T4 is turbine wheel disc radius run-out degree.

Fig. 4 is that gas turbine rotor unbalanced moment produces schematic diagram.

Fig. 5 is compressor end bearing place axle amplitude before and after gas turbine pull rod rotor unbalance phase optimization.

Fig. 6 is the turbine stub shaft place of holding axle amplitude before and after gas turbine pull rod rotor unbalance phase optimization.

Fig. 7 is gas turbine pull rod rotor unbalance phase optimization rear bearing place axle amplitude.

Fig. 8 is the structural representation of end-tooth.

Detailed description of the invention:

Below in conjunction with accompanying drawing, the present invention is described in further detail.

Fig. 1 is certain combustion engine center pull rod rotor typical structure schematic diagram, and rotor wheel disc at different levels relies on end-tooth torsion pass, adopts center pull rod to apply pretightning force.

Referring to Fig. 2 to Fig. 8, the present invention, with the gas turbine pull rod rotor dynamic balancing ability Optimization Design of end face tooth structure, comprises the following steps:

1) determination of amount of unbalance suffered by gas turbine rotor.

For obtaining the data of each parts amount of unbalance of whole rotor, according to manufacturer provide gas turbine rotor before and after spindle nose, wheel disc at different levels radial run-out degree obtain eccentric throw and the unbalance phase of each parts.Wherein the relation of eccentric throw δ and radial run-out degree e is；Result according to eccentric throw and unbalance phase, determine the distribution of whole rotor initial unbalance, the each parts of whole rotor have 25 places needs to add the position of amount of unbalance, relation between amount of unbalance and eccentric throw is: q=m δ, and then obtains the relation between amount of unbalance and radius run-out degree and be:Determining each parts amount of unbalance of rotor, wherein, q is amount of unbalance, and e is radial run-out degree, and m is this grade of wheel disc quality.Fig. 2 gives the schematic diagram of wheel disc amount of unbalance, and it produces amount of unbalance component in x direction and y direction.Fig. 3 gives the distribution of whole rotor unbalance value, has 25 place's amount of unbalances and is applied to rotor position.

2) determination of gas turbine rotor imbalance centrifugal force and out-of-balance force moment of flexure optimization object function.

The amount of unbalance amount of unbalance component in x direction and y direction can produce the uneven centrifugal force F of respective direction_x, F_yWith out-of-balance force moment M_x, M_y, Fig. 4 provides the mechanism of production of out-of-balance force moment of flexure.Uneven centrifugal force and out-of-balance force Moment make rotor produce vibration, according to uneven centrifugal force and out-of-balance force moment of flexure mechanism of production, it is determined that uneven centrifugal force and out-of-balance force moment of flexure optimization object function be:

The vector of uneven centrifugal force and moment of flexure is represented by:

Wherein, q (i) represents the amount of unbalance/g mm of wheel disc at different levels；α (i) represent amount of unbalance phase place/°, for the variable in genetic algorithm；K represents wheel disc progression；L (i) represents the wheel disc amount of unbalance at different levels distance/mm to rotor midpoint position；ω is rotor speed/r min^-1；I span is 1～25；F_xFor x direction imbalance centrifugal force/N；F_yFor y direction imbalance centrifugal force/N；F is total uneven centrifugal force/N；M_xFor x direction out-of-balance force moment of flexure/N m；M_yFor y direction out-of-balance force moment of flexure/N m；M is total out-of-balance force moment of flexure/N m.

3) determination of the wheel disc established angle of the minima of uneven centrifugal force and moment of flexure and correspondence thereof.

According to uneven centrifugal force and out-of-balance force moment of flexure optimization object function, working out genetic algorithm optimization program, wherein α (i) is the variable in optimized algorithm.By adjusting the established angle of wheel disc at different levels, it is determined that the minima of object function, i.e. the minima of uneven centrifugal force and moment of flexure, the setting angle of its correspondence is the established angle of needs.Needs particularly point out, and for the special construction that end-tooth connects, owing to the whole circle of wheel disc end-tooth has 180 teeth, therefore the established angle calculated are rounded and ensure that the setting angle difference of adjacent wheel disc is even number.

4) unbalance response of contrast initial settling angle degree and optimization setting angle lower rotor part.

The reliability of wheel disc setting angle is optimized for checking, adopt rotor unbalance method of response calculation that unbalance response amplitude before and after optimizing is contrasted, after optimization, the unbalance response amplitude of rotor reduces, the maximum amplitude that rotor allows when the axle amplitude at bearing place is less than dynamic balancing.

After optimization established angle carry out unbalance responses and with optimize before unbalance responses Comparative result.Compressor end bearing place axle amplitude before and after optimizing is as it is shown in figure 5, the first-order kernel peak value after optimizing drops to 0.802 μm from 482 μm, and fall is more than 95%；Second-order response peak value drops to 2.91 μm from 60.90 μm before optimization, and fall is more than 95%.

Optimizing forward and backward turbine stub shaft and hold place's axle amplitude as shown in Figure 6, after optimization, first-order kernel peak value drops to 0.843 μm from 521 μm, and fall is more than 95%；Second-order response peak value drops to 1.88 μm from 9.48 μm before optimization.Fall is more than 80%.

Compressor end and the turbine stub shaft place of holding axle amplitude after optimization are as it is shown in fig. 7, first-order kernel peak-fall after optimizing, and the axle amplitude of turbine end and compressor end is respectively less than 1 μm；Second-order response peak value after optimization relatively optimizes front decline, and the axle amplitude of turbine end and compressor end is respectively less than 3 μm；When working speed is 3000r/min, the axle amplitude of turbine end and compressor end is respectively less than 2.3 μm.

Before and after table 1 gas turbine rotor optimization, unbalance response axle shakes single magnitude determinations Comparative result analysis

Claims (4)

1. the gas turbine pull rod rotor high-speed balancing ability Optimization Design with end-tooth, it is characterised in that comprise the following steps:

1) the radial run-out value e according to the front spindle nose of gas turbine pull rod rotor, rear spindle nose and wheel disc at different levels, it is determined that the initial size of whole rotor difference parts amount of unbalance and PHASE DISTRIBUTION；

2) according to the initial size of amount of unbalance and PHASE DISTRIBUTION, it is determined that uneven centrifugal force to be optimized and each uneven centrifugal force are object function to the moment of flexure of rotor midpoint；

3) work out genetic algorithm optimization program according to object function, adopt genetic algorithm to obtain the minima of object function and the wheel disc end-tooth setting angle of correspondence thereof；

4) by contrasting initial settling angle degree and optimizing the unbalance response of setting angle lower rotor part, after optimization, axle amplitude first-order kernel peak-fall amplitude in compressor end bearing place is more than 95%, and turbine stub shaft holds place's axle amplitude first-order kernel peak-fall amplitude more than 95%；After optimization, axle amplitude second-order response peak-fall amplitude in compressor end bearing place is more than 95%, and turbine stub shaft holds place's axle amplitude second-order response peak-fall amplitude more than 80%, it is determined that the genetic algorithm effectiveness to gas turbine rotor high-speed balancing.

2. the gas turbine pull rod rotor high-speed balancing ability Optimization Design with end-tooth according to claim 1, it is characterized in that, described gas turbine pull rod rotor, with end face tooth structure, is connected by end-tooth between adjacent wheel disc, and end-tooth attachment structure is used for adjusting wheel disc setting angle.

3. the gas turbine pull rod rotor high-speed balancing ability Optimization Design with end-tooth according to claim 1, it is characterised in that step 1) in, the relation between radial run-out degree e and amount of unbalance is:

$q=\frac{e\·m}{2}$

Wherein, q is amount of unbalance, and e is radial run-out degree, and m is this grade of wheel disc quality.

4. the gas turbine pull rod rotor high-speed balancing ability Optimization Design with end-tooth according to claim 1, it is characterised in that step 2) in, optimization object function is:

$\left\{\begin{array}{c}{F}_x=\underset{i=1}{\overset{k}{\Σ}}q\left(i\right)\×cos\left(\mathrm{\α}\right(i))\×{\mathrm{\ω}}^2\\ F_y=\underset{i=1}{\overset{k}{\Σ}}q\left(i\right)\×sin\left(\mathrm{\α}\right(i))\×{\mathrm{\ω}}^2\\ M_x=\underset{i=1}{\overset{k}{\Σ}}L\left(i\right)\×q\left(i\right)\×\sin \left(\mathrm{\α}\right(i))\×{\mathrm{\ω}}^2\\ M_y=\underset{i=1}{\overset{k}{\Σ}}L\left(i\right)\×q\left(i\right)\×\cos \left(\mathrm{\α}\right(i))\×{\mathrm{\ω}}^2\end{array}\right.$

The vector of uneven centrifugal force and out-of-balance force moment of flexure is expressed as:

$\left\{\begin{array}{c}F=\sqrt{F_x^2+F_y^2}\\ M=\sqrt{M_x^2+M_y^2}\end{array}\right.$

Wherein, q (i) represents the amount of unbalance/g mm of wheel disc at different levels；α (i) represent amount of unbalance phase place/°, for the variable in genetic algorithm, namely corresponding wheel disc setting angle；K represents wheel disc progression；ω is rotor speed/r min^-1；L (i) represents the wheel disc amount of unbalance at different levels distance/mm to rotor midpoint position；I span is 1～25；F_xFor x direction imbalance centrifugal force/N；F_yFor y direction imbalance centrifugal force/N；F is total uneven centrifugal force/N；M_xFor x direction out-of-balance force moment of flexure/N m；M_yFor y direction out-of-balance force moment of flexure/N m；M is total out-of-balance force moment of flexure/N m.