CN113190933B - Load reproducing and extrapolation method for internal combustion engine of engineering machinery - Google Patents
Load reproducing and extrapolation method for internal combustion engine of engineering machinery Download PDFInfo
- Publication number
- CN113190933B CN113190933B CN202110620349.5A CN202110620349A CN113190933B CN 113190933 B CN113190933 B CN 113190933B CN 202110620349 A CN202110620349 A CN 202110620349A CN 113190933 B CN113190933 B CN 113190933B
- Authority
- CN
- China
- Prior art keywords
- load
- torque
- rotating speed
- interval
- extrapolation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000013213 extrapolation Methods 0.000 title claims abstract description 76
- 238000002485 combustion reaction Methods 0.000 title claims abstract description 44
- 238000000034 method Methods 0.000 title claims abstract description 40
- 230000001360 synchronised effect Effects 0.000 claims abstract description 30
- 238000012952 Resampling Methods 0.000 claims abstract description 29
- 238000005070 sampling Methods 0.000 claims description 18
- 239000011159 matrix material Substances 0.000 claims description 14
- 238000010586 diagram Methods 0.000 claims description 13
- 230000001186 cumulative effect Effects 0.000 claims description 11
- 238000010276 construction Methods 0.000 claims description 10
- 230000008859 change Effects 0.000 claims description 6
- 238000005315 distribution function Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 238000009825 accumulation Methods 0.000 claims description 4
- 230000008030 elimination Effects 0.000 claims description 4
- 238000003379 elimination reaction Methods 0.000 claims description 4
- 230000007774 longterm Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000000295 complement effect Effects 0.000 claims description 3
- 230000001172 regenerating effect Effects 0.000 claims 1
- 238000001228 spectrum Methods 0.000 abstract description 7
- 238000000605 extraction Methods 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06M—COUNTING MECHANISMS; COUNTING OF OBJECTS NOT OTHERWISE PROVIDED FOR
- G06M1/00—Design features of general application
- G06M1/27—Design features of general application for representing the result of count in the form of electric signals, e.g. by sensing markings on the counter drum
- G06M1/276—Design features of general application for representing the result of count in the form of electric signals, e.g. by sensing markings on the counter drum using mechanically-actuated contacts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/12—Timing analysis or timing optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Testing Of Engines (AREA)
Abstract
The invention discloses a method for reproducing and extrapolating the load of an internal combustion engine of engineering machinery, belonging to the technical field of load spectrum compilation of the internal combustion engine of the engineering machinery, comprising the following steps: (1) extracting torque and rotation speed synchronous peak-valley values; (2) synchronously eliminating small loads by torque and rotating speed; (3) resampling torque and rotating speed data; (4) synchronously counting torque and rotating speed data; (5) selecting a threshold value; (6) proportionally extrapolating the middle and low load intervals; (7) extreme value extrapolation of the POT model; (8) and synchronously counting the load after the extrapolation of the extreme value. The method solves the problem that the torque and rotating speed double-parameter load of the internal combustion engine is difficult to synchronously reproduce in the typical operation stage of the engineering machinery, proportionally extrapolates the low and medium loads of the internal combustion engine, and extrapolates the extreme value of the POT model of the extreme value load, so that the reproducing and extrapolating precision of the extreme value load of the internal combustion engine is improved, and the method is simple and applicable.
Description
Technical Field
The invention relates to the technical field of load spectrum compilation of an internal combustion engine of engineering machinery, in particular to a method for reproducing and extrapolating the load of the internal combustion engine of the engineering machinery.
Background
The engineering machinery (such as a loader and an excavator) has the characteristics of severe environment, variable working conditions, long continuous working time and the like in the working process, which causes great challenges to the reliability and durability of an internal combustion engine of the engineering machinery, so that the load of the internal combustion engine of the engineering machinery is reproduced and the full-life extrapolation is carried out, which is of great significance to the research of the internal combustion engine of the engineering machinery, and the reproduced load of the internal combustion engine can also be used for the reliability analysis of parts of the internal combustion engine, for example, the fatigue analysis of a crankshaft, a connecting rod, a cylinder body, a piston and the like is carried out based on the complete machine load spectrum of the internal combustion engine.
The load of an internal combustion engine of a construction machine is generally composed of a plurality of parameters, wherein the parameters which have a large influence on the internal combustion engine are torque and rotating speed, and the torque and the rotating speed can reflect the loaded force and the loaded frequency when the internal combustion engine works. However, the research literature for simultaneously reproducing and extrapolating the load of two load parameters is relatively few, and how to simultaneously reproduce and extrapolate the rotating speed and the torque load of the internal combustion engine is an important subject of reliability analysis of the internal combustion engine of the engineering machinery. The patent publication No. CN103678256A, the inventor's luo qing dynasty patent "load spectrum compilation method for vehicle engine" compiles a load spectrum of vehicle engine by calculating a load transition time matrix and a load retention time matrix, but the method does not consider the influence of extreme load in the process of load extrapolation, so the compiled load spectrum is conservative.
In conclusion, the torque and rotational speed loads of the internal combustion engine of the construction machine are reproduced and extrapolated at the same time, and the influence of extreme loads is considered in the extrapolation process, which is the direction in which the load reproduction and extrapolation of the internal combustion engine of the construction machine needs to be studied urgently.
Disclosure of Invention
The invention aims to provide a method for reproducing and extrapolating the load of an internal combustion engine of engineering machinery, which aims to solve the problems that the rotation speed and the torque load of the internal combustion engine are reproduced simultaneously and the extreme load is difficult to extrapolate.
In order to achieve the purpose, the invention adopts the technical scheme that:
the method for reproducing and extrapolating the load of the internal combustion engine of the engineering machinery comprises the following steps:
(1) torque and rotating speed synchronous peak-valley value extraction
The peak-to-valley data represents the turning of the load and contains important information of the load. When the peak-valley values of the torque and the rotating speed are extracted, the data of the time corresponding to the peak-valley values of the torque and the rotating speed are extracted at the same time, and the method comprises the following specific steps:
(1.1) extracting the peak-to-valley value of the torque, and recording a time sequence matrix Tt T corresponding to the peak-to-valley value of the torquet＝[T1,T2,T3......Tn]；
(1.2) extracting the peak-to-valley value of the rotating speed, and recording a time sequence matrix Tr corresponding to the peak-to-valley value of the rotating speed
Tr＝[T1',T2',T3'......Tn']；
(1.3) combining the time-series matrices Tt and Tr into an overall peak-to-valley time-series matrix of [ T ═ T [ ]1,T1',T2,T2',T3,T3'......Tn,Tn']；
(1.4) extracting rotating speed or torque sample points acquired at corresponding time according to the total time sequence matrix to obtain a rotating speed or torque peak-valley time sequence simultaneously containing rotating speed and torque peak-valley time;
(2) torque and rotating speed synchronous eliminating small load
When carrying out little load and rejecting, in order to guarantee that the data point moment of moment and rotational speed keeps unanimous, carry out the rejection of little load to moment, rotational speed simultaneously, concrete step is:
the absolute difference between the load value corresponding to the torque at a certain moment and the load values corresponding to the two moments before and after the moment is less than a threshold value Delta1Meanwhile, the absolute difference value between the rotating speed load value corresponding to the moment and the rotating speed load values corresponding to the front moment and the rear moment is smaller than a threshold value Delta2At the moment, the torque and rotating speed data values at the moment are simultaneously rejected;
(3) torque and rotational speed data resampling
(3.1) in order to conveniently and later perform synchronous counting and improve the counting precision, the time step between two data points is as short as possible and is equally spaced, so that the torque and the rotating speed are re-sampled, the sampling rate is improved, and the time step between the two load data points before and after is shortened;
in order to increase the resampling rate to an integral multiple of the original sampling frequency when more load data points fall on the real collected load data points during resampling, the time step Δ t between the two data points after resampling is as follows:
△t＝1/mf (1)
wherein f is the original sampling frequency; m is the ratio of the resampling frequency to the original sampling frequency;
(3.2) because the sampling frequency is improved, some data points are empty during resampling, and at this time, an interpolation method is adopted to complement sample points with empty data, that is, if a data point at a certain moment is empty during resampling, two points are interpolated once by using two adjacent data points at the moment, and the formula is as follows:
wherein xiThe moment when the data point is empty; y isiThe data point value after the linear interpolation at the moment is taken as the data point value; x is the number ofi+1The latter moment; y isi+1The data point value at the later moment; x is the number ofi-1The previous moment; y isi-1The value of the data point at the previous time.
(4) Synchronous counting of torque and rotating speed data
The method comprises the following steps of synchronously counting torque and rotating speed data, and counting the interval where the collected torque and rotating speed data are located and the accumulated time of the interval, wherein the method comprises the following specific steps:
(4.1) firstly, carrying out scatter diagram drawing on the torque and rotating speed data, finding respective interval ranges of the torque and rotating speed data, and then dividing torque and rotating speed intervals;
(4.2) carrying out interval statistics on the torque and the rotating speed: finding an interval where a torque and rotating speed load value corresponding to a certain moment are located, accumulating the time of the interval once, wherein the accumulated time is equal to the time of the interval and adding a time step delta t after resampling;
(5) threshold selection
Before extrapolation of torque and rotating speed load, threshold values of the torque and rotating speed load are selected, extreme value load above the threshold values is extrapolated by using an extreme value of a POT (point of arrival) model, and medium and low load below the threshold values is extrapolated by using proportion.
The threshold selection principle of the rotating speed and torque data is the same: the extreme sample volume produced at this threshold being no greater than the total collected sample volume
Namely:
wherein n isuIs the extreme sample capacity; n is the total collected sample volume;
the value of (a) is determined by the intensity of the load change;
(6) intermediate and low load interval proportional extrapolation
Determining threshold values of the torque and the rotating speed from the step (5), when the interval of the torque and the rotating speed in the step (4) is below the respective threshold value, indicating that the load of the interval is the medium-low load, directly carrying out proportional extrapolation on the accumulated time of the interval of the medium-low load counted in the step (4), wherein a proportional extrapolation coefficient k is as follows:
k＝T×w％/t (4)
wherein T is the expected life of the internal combustion engine of the engineering machinery; w% is the ratio of the working condition; t is the time to collect the sample load;
(7) extreme extrapolation of POT model
For the condition that the accumulated time of the interval where the extreme load is located cannot be directly proportionally extrapolated like the step (6), POT model extreme value extrapolation is firstly carried out, and then synchronous counting of torque and rotating speed is carried out on the extrapolated load time sequence, wherein the POT model extrapolation specifically comprises the following steps:
(7.1) Generalized Pareto Distribution (GPD) fitting extreme loads
The cumulative distribution function expression for GPD is:
wherein σ is a scale parameter; xi is a shape parameter; u is a threshold value;
(7.2) extreme extrapolation
When the POT model is used for extrapolation, only the extreme load is extrapolated, the extreme value after extrapolation is the result of increasing the GPD probability corresponding to the original extreme load by alpha%, and the alpha value is determined by the intensity of load change under different working conditions.
(8) Load synchronous counting after extremum extrapolation
When the torque and the rotating speed are synchronously counted for the long-term load time sequence after the pole value extrapolation, in order to save the time for synchronous counting, all load data points after the extrapolation do not need to be synchronously counted, only the interval accumulation time of the interval where the extreme load above the threshold is located needs to be counted again, and the accumulation time of the interval where the middle and low loads are located adopts the result of the proportional extrapolation in the step (6).
Further, in the step (2), the torque small load rejection threshold value delta1And speed of rotation small load rejection threshold delta2The calculation formula of (2) is as follows:
△1or delta2＝(△max-△min)×θ×2 (6)
Wherein, ΔmaxIs the maximum amplitude of the torque or rotational speed load cycle collected; deltaminIs the minimum amplitude of the torque or rotational speed load cycle collected; theta is a small load rejection threshold;
in the step (4), when the torque and the load are divided into the interval ranges, the interval ranges are slightly larger and the interval ranges are slightly smaller according to the divided intervals in the scatter diagram, wherein the load concentration is unequal, the divided intervals are not the places where the load is concentrated, and the divided intervals are the places where the load is concentrated.
Further, in the step (8), when the torque and the rotation speed are synchronously counted for the long-term load time sequence after the pole value extrapolation, it is not necessary to count all load data points synchronously, and it is only necessary to count the accumulated time again for the interval where the extreme load above the threshold is located, where the extreme load interval above the threshold includes any one of the following three cases:
a. a section above a torque threshold;
b. an interval above a rotational speed threshold;
c. while in the interval above the torque and speed thresholds.
The invention has the beneficial effects that:
(1) the invention can simultaneously eliminate small-amplitude loads of torque and rotating speed, ensure that time sequence data points of the torque and the rotating speed are always in one-to-one correspondence, and truly reproduce the torque and rotating speed load spectrum of the internal combustion engine under a specific working condition.
(2) The invention shortens the time step during synchronous counting by resampling and linear interpolation, and carries out scatter diagram drawing before torque and rotating speed interval division to find out a proper interval division range, thereby improving the precision of synchronous counting.
(3) When load extrapolation is carried out, the method directly carries out proportional extrapolation of interval accumulated time for medium and low loads, firstly utilizes a POT model to extrapolate an extreme value for extreme value loads, and then synchronously counts torque and rotating speed loads above a threshold value again. The method combines the proportion extrapolation and the extreme value extrapolation of the POT model, shortens the time for synchronously counting the extrapolated long-term load sequence, considers the influence of the extreme value load, is simple and reliable, is more suitable for engineering practice, and has wide engineering application value.
Drawings
FIG. 1 is a flow chart of load reproduction and extrapolation for an internal combustion engine of a construction machine;
FIG. 2 is a flow chart of torque and rotation speed peak-valley synchronous extraction;
FIG. 3 is a schematic diagram of torque and rotation speed synchronous peak-to-valley extraction;
FIG. 4 is a schematic diagram of torque and rotation speed synchronous elimination of small loads;
FIG. 5a is a line graph of torque load data before small load rejection;
FIG. 5b is a line graph of torque load data after a small load rejection;
FIG. 6a is a data point plot before torque load interpolation;
FIG. 6b is a plot of a post torque load interpolation data point;
FIG. 7 is a scatter plot of rotational speed, torque load;
FIG. 8 is a two-dimensional plot of torque, speed, and cumulative time for different intervals;
FIG. 9a is a plot of extreme loads with torque above a threshold;
FIG. 9b is an extreme load plot with speed above threshold;
FIG. 10 is a two-dimensional plot of torque, rotational speed, and accumulated time after proportional extrapolation;
FIG. 11a is a graph of a cumulative distribution function of extreme torque loads;
FIG. 11b is a graph of a cumulative distribution function of the load at the extreme rotational speed;
FIG. 12 is a schematic representation of torque load extremum extrapolation;
FIG. 13 is a schematic representation of extreme load extrapolation and synchronization count.
Detailed Description
The invention discloses a method for reproducing and extrapolating load of an internal combustion engine of engineering machinery.
The present invention will be further described with reference to the following examples and drawings, but the present invention is not limited thereto.
The invention provides a method for reproducing and extrapolating the load of an internal combustion engine of engineering machinery, which takes a certain excavator as an example, and collects the load samples of the internal combustion engine with the typical operation duration of one hour, namely the time sequence of torque and rotating speed load, and the sampling frequency is 1 Hz. The process of load reproduction and extrapolation of the internal combustion engine of the engineering machinery is shown in the figure 1 by taking the reproduction and extrapolation of the load of the torque and the rotating speed as an example, and comprises the following steps:
(1) torque and rotation speed synchronous peak-valley value extraction
The peak-to-valley data represents the turning of the load and contains important information of the load. When extracting the peak-to-valley values of the torque and the rotating speed, data of time corresponding to the peak-to-valley values of the torque and the rotating speed are extracted at the same time, and the specific steps are shown in the flow chart of fig. 2:
(1.1) extracting the peak-to-valley value of the torque, and recording a time sequence matrix T corresponding to the peak-to-valley value of the torquet＝[T1,T2,T3......Tn]；
(1.2) extracting the peak-to-valley value of the rotating speed, and recording a time sequence matrix T corresponding to the peak-to-valley value of the rotating speedr＝[T1',T2',T3'......Tn']；
(1.3) combining the time-series matrices Tt and Tr into an overall peak-to-valley time-series matrix of [ T ═ T [ ]1,T1',T2,T2',T3,T3'......Tn,Tn']；
(1.4) extracting rotating speed or torque sample points acquired at corresponding time according to the total time sequence matrix to obtain a rotating speed or torque peak-valley time sequence simultaneously containing rotating speed and torque peak-valley time;
fig. 3 is a schematic diagram of a section of captured torque and rotation speed after peak-valley value extraction, in which the dotted line is a torque and rotation speed load time sequence after peak-valley value extraction, and the torque and rotation speed load time sequence at this time includes all data points of time corresponding to the torque and rotation speed peak-valley value.
(2) Torque and rotating speed synchronous eliminating small load
When small load rejection is performed, in order to ensure that data points of torque and rotation speed are kept consistent all the time, small load rejection is performed on the torque and the rotation speed at the same time, and the specific steps are as shown in a schematic diagram of fig. 4:
the absolute difference between the load value corresponding to the torque at the time t1 and the load values corresponding to the two preceding and following times is less than the threshold Δ1Meanwhile, the absolute difference value between the rotating speed load value corresponding to the moment and the rotating speed load values corresponding to the front moment and the rear moment is smaller than the threshold delta2At the moment, the data values of the torque and the rotating speed at the moment are simultaneously rejected;
torque small load elimination threshold delta1And the rotation speed and small load rejection threshold delta2The formula (2) is shown in formula (1):
△1＝(△max-△min)×θ×2 (1)
wherein, ΔmaxIs the maximum amplitude of the torque load cycle collected; deltaminIs the minimum amplitude of the torque load cycle collected; theta is a small load rejection threshold;
in this embodiment, 3600 sample points are collected altogether, 440 small loads are removed, and 3160 effective data points are finally obtained, and fig. 5a and 5b show that the number of sample points is reduced after the small loads are removed from the collected torque load data, and similarly, 3160 effective sample points are obtained after the small loads are removed from the rotational speed load samples.
(3) Torque and rotational speed data resampling
(3.1) in order to facilitate the subsequent synchronous counting and improve the counting precision, the time step between two data points is as short as possible and at equal intervals, so that the torque and the rotating speed are resampled, the sampling rate is improved, and the time step between the two load data points is shortened;
in order to increase the resampling rate to an integral multiple of the original sampling frequency in order to fall on more actually collected load data points during resampling, in this embodiment, 2 times of the original sampling frequency is used as the resampling frequency, and a time step Δ t between two data points before and after resampling is:
△t＝1/2f (2)
wherein, f is the original sampling frequency;
in this embodiment, the original sampling frequency of the torque and the rotation speed is 1Hz, so that the initial time step between data points is 1 second, and after resampling, the sampling frequency is 2 Hz.
(3.2) because the sampling frequency is improved, some data points are empty during resampling, and at this time, an interpolation method is adopted to complement sample points with empty data, that is, if a data point at a certain moment is empty during resampling, two points are interpolated once by using two adjacent data points at the moment, and the formula is as follows:
wherein xiThe moment when the data point is empty; y isiThe data point value after the linear interpolation at the moment is taken as the data point value; x is the number ofi+1The latter moment; y isi+1The data point value at the next moment; x is the number ofi-1The previous moment; y isi-1The value of the data point at the previous time.
Comparing the sampled torque load segments of fig. 6a and 6b before and after resampling and interpolation, it can be seen that 30 data points are obtained before linear interpolation, and the number of data points is doubled and changed to 60 data points within the same acquisition time after linear interpolation, so the time step between the two data points is changed from the original 1 second to 0.5 second, and the time step is shortened.
(4) Synchronous counting of torque and rotating speed data
The method comprises the following steps of synchronously counting torque and rotating speed data, and counting the interval where the collected torque and rotating speed data are located and the accumulated time of the interval, wherein the method comprises the following specific steps:
(4.1) firstly, carrying out scatter diagram drawing on the torque and rotating speed data, finding respective interval ranges of the torque and rotating speed data, and then dividing torque and rotating speed intervals;
FIG. 7 is a drawing result of a scatter diagram of collected rotational speed and torque load data, and it can be seen that the torque is mainly concentrated in the range of [ -50Nm to 250Nm ], but is particularly concentrated in the range of [0 to 160Nm ], so that the torque is divided into [ -50 to 0], [0 to 40], [40 to 80], [80 to 120], [120 to 160], [160 to 250 ].
As can be seen from FIG. 7, the rotation speed range is mainly concentrated in the range of [1150r/min to 1500r/min ], especially in the range of [1300r/min to 1450r/min ], so the rotation speed range is divided into [1150 to 1300], [1300 to 1350 to 1400], [1400 to 1450], and [1450 to 1500 ].
(4.2) carrying out interval statistics on the torque and the rotating speed: finding an interval where a torque and a rotating speed load value corresponding to a certain moment are located, and accumulating the time of the interval once, wherein the accumulated time is equal to the time of the interval and is added with a time step delta t after resampling, and the delta t is 0.5 second in the embodiment;
fig. 8 is a two-dimensional graph of torque, rotation speed and accumulated time in different sections obtained after synchronous counting and division of collected data of torque and rotation speed with sample length of one hour, wherein the total time in the graph is 3160 seconds, which is the remaining effective time after small loads are removed, and synchronous counting is performed by 3160 × 2-6320 steps in total, with each interval of 0.5 seconds.
(5) Threshold selection
Before extrapolation of torque and rotating speed load, threshold values of the torque and rotating speed load are selected, extreme value load above the threshold values is extrapolated by using an extreme value of a POT (point of arrival) model, and medium and low load below the threshold values is extrapolated by using proportion.
The threshold selection principle of the rotating speed and torque data is the same: the extreme sample volume produced at this threshold being no greater than the total collected sample volume
Namely:
wherein n isuIs the extreme sample capacity; n is the total collected sample volume;
the value of (a) is determined by the intensity of the load change;
in the present embodiment, the first and second electrodes are,
the value is 10%, 3600 torque load points and 3600 rotating speed load points are collected, so when the extreme load quantity of the torque and the rotating speed is 360, the corresponding load value is the threshold value of the rotating speed or the torque. FIGS. 9a and 9b are extreme load graphs with a section of captured torque and rotational speed data above their respective thresholds, respectively, in this embodiment, the torque threshold is 139.5Nm, and the rotational speed threshold is 1391.2 r/min.
(6) Intermediate and low load interval proportional extrapolation
Determining threshold values of the torque and the rotating speed from the step (5), when the interval of the torque and the rotating speed in the step (4) is below the respective threshold value, indicating that the load of the interval is the medium-low load, directly carrying out proportional extrapolation on the accumulated time of the interval of the medium-low load counted in the step (4), wherein a proportional extrapolation coefficient k is as follows:
k＝T×w％/t (5)
wherein T is the expected life of the internal combustion engine of the engineering machinery; w% is the ratio of the working condition; t is the time to collect the sample load;
in the present embodiment, in order to simplify the calculation, 5-fold proportional extrapolation is performed on the torque, the rotation speed, and the accumulated time interval, that is, the proportional extrapolation coefficient k is 5. Fig. 10 is a two-dimensional graph of torque, rotation speed and cumulative time after exceptional ratio extrapolation, wherein a gray background interval is an interval in which the medium and low load is located, as can be seen from the graph, the cumulative time of each torque and load interval is increased by 5 times, and the total time is increased from 3160 seconds to 15800 seconds.
(7) Extreme extrapolation of POT model
For the condition that the accumulated time of the interval where the extreme load is located cannot be directly proportionally extrapolated like the step (6), firstly carrying out POT model extreme value extrapolation to obtain extrapolated load, and then carrying out synchronous counting on torque and rotating speed, wherein the POT model extrapolation specifically comprises the following steps:
(7.1) Generalized Pareto Distribution (GPD) fitting extreme loads
The cumulative distribution function expression for GPD is:
wherein σ is a scale parameter; xi is a shape parameter; u is a threshold;
the torque threshold calculated in step (5) is 139.5Nm, the rotational speed threshold is 1391.2r/min, the torque and the extreme load of the rotational speed above the threshold are respectively fitted by two GPD functions, the obtained cumulative distribution functions are shown in fig. 11a and 11b, and the fitting parameters are shown in table 1.
TABLE 1GPD function fitting parameters
(7.2) extreme extrapolation
The extreme value extrapolation of the POT model is to extrapolate only the extreme load, the extrapolated extreme value is the result of the probability improvement of alpha% of the original GPD, and the alpha value is determined by the intensity of the load change under different working conditions.
In this embodiment, α is 5, the extrapolation coefficient k is 5, the load time series after 5 times of the POT model extrapolation is connected end to end, fig. 12 is a torque load schematic diagram after one small section of the intercepted extreme value is extrapolated, it can be seen from the figure that the dotted line is the load after the extreme value extrapolation, only the extreme value load above the threshold value is extrapolated, and the medium and low loads are directly reproduced.
(8) Load synchronous counting after extremum extrapolation
When carrying out the synchronous count of moment of torsion and rotational speed to the load time sequence after the extreme extrapolates, in order to save count time, need not all carry out the synchronous count to whole load data points, only need carry out the synchronous count again to the extreme load more than the threshold value, count the accumulative total time of extreme load place interval, extreme load place interval includes following three kinds of arbitrary condition:
a. an interval above a torque threshold;
b. an interval above a rotational speed threshold;
c. while in the interval above the torque and speed thresholds.
FIG. 13 is a diagram showing that only extreme load synchronous counting is performed on the obtained load time series after 5 times of extrapolation of extreme load by the POT model;
it can be seen from the figure that the cumulative time of the interval with the gray background is the same as the cumulative time of the same gray background interval of figure 10, which is the result of the proportional extrapolation. For the interval where the extreme load is located, the accumulated time obtained by using two different extrapolation methods, namely proportional extrapolation and POT model extreme extrapolation, is different, and the POT model extreme extrapolation takes the diversity of the extreme load in work into consideration and introduces the idea of random distribution, so that the distribution of the extreme load interval is more dispersed and accords with the reality.
According to the method for reproducing and extrapolating the load of the internal combustion engine of the engineering machinery, the time for synchronously counting the extrapolated load is shortened by extrapolating the middle and low loads in proportion, the extreme load is extrapolated by extrapolating the extreme load by using the POT model, the extreme load diversity is considered, the proportion extrapolation and the POT model extreme extrapolation are combined, the method is simple and reliable, is more suitable for engineering practice, provides a basis for the service life research of the internal combustion engine of the engineering machinery, the reliability analysis of the whole machine and the fatigue analysis of parts of the internal combustion engine, and has wide engineering application value.
Claims (7)
1. The method for reproducing and extrapolating the load of the internal combustion engine of the engineering machinery is characterized by comprising the following steps:
(1) extracting torque and rotating speed synchronous peak-valley values;
(2) synchronously eliminating small loads by torque and rotating speed;
(3) resampling torque and rotating speed data;
(4) synchronously counting torque and rotating speed data;
(5) selecting a threshold value;
(6) extrapolation of the interval proportion of the medium and low loads;
(7) extreme value extrapolation of the POT model;
(8) synchronously counting the load after the extremum is extrapolated;
in the step (4), in the synchronous counting of the torque and rotating speed data, the torque and rotating speed data are synchronously counted, and the interval where the collected torque and rotating speed data are located and the accumulated time of the interval are counted, wherein the method specifically comprises the following steps:
(4.1) firstly, carrying out scatter diagram drawing on the torque and rotating speed data, finding respective interval ranges of the torque and rotating speed data, and then dividing torque and rotating speed intervals;
when the torque and the load are divided into the interval ranges, dividing the interval with unequal intervals according to the concentration condition of the load in the scatter diagram, wherein the load is not concentrated, the divided interval range is a little bit larger, and the load is concentrated, and the interval range is a little bit smaller;
(4.2) carrying out interval statistics on the torque and the rotating speed: finding an interval where a torque and rotating speed load value corresponding to a certain moment are located, accumulating the time of the interval once, wherein the accumulated time is equal to the time of the interval and adding a time step delta t after resampling;
in the step (8), when the torque and the rotating speed are synchronously counted for the long-term load time sequence after the pole value extrapolation, in order to save the time for synchronous counting, all load data points after the extrapolation do not need to be synchronously counted, only the interval accumulation time of the interval where the extreme load above the threshold is located needs to be counted again, and the accumulation time of the interval where the medium and low loads are located adopts the result of the proportional extrapolation in the step (6); the extreme load interval above the threshold includes any one of the following three cases:
a. a section above a torque threshold;
b. an interval above a rotational speed threshold;
c. while in the interval above the torque and speed thresholds.
2. The method for load reproduction and extrapolation for internal combustion engine of construction machinery according to claim 1,
in the step (1), in the process of extracting the peak-valley values of the torque and the rotating speed synchronously, when the peak-valley values of the torque and the rotating speed are extracted, data of time corresponding to the peak-valley values of the torque and the rotating speed are extracted simultaneously, and the method specifically comprises the following steps:
(1.1) extracting the peak-to-valley value of the torque, and recording a time sequence matrix Tt T corresponding to the peak-to-valley value of the torquet＝[T1,T2,T3......Tn]；
(1.2) extracting the peak-to-valley value of the rotating speed, and recording a time sequence matrix Tr corresponding to the peak-to-valley value of the rotating speed
Tr＝[T1',T2',T3'......Tn']；
(1.3) combining the time-series matrices Tt and Tr into an overall peak-to-valley time-series matrix of [ T ═ T [ ]1,T1',T2,T2',T3,T3'......Tn,Tn']；
And (1.4) extracting rotating speed or torque sample points acquired at corresponding time according to the total time sequence matrix to obtain a rotating speed or torque peak-valley time sequence simultaneously containing the rotating speed and the torque peak-valley time.
3. The method for load reproduction and extrapolation for internal combustion engine of construction machinery according to claim 1,
in the step (2), in the synchronous elimination of the torque and the rotating speed of the small load, when the small load is eliminated, in order to ensure that data points of the torque and the rotating speed are kept consistent all the time, the small load is eliminated simultaneously on the torque and the rotating speed, and the method comprises the following specific steps:
the absolute difference between the load value corresponding to the torque at a certain moment and the load values corresponding to the two moments before and after the moment is less than a threshold value Delta1Meanwhile, the absolute difference value between the rotating speed load value corresponding to the moment and the rotating speed load values corresponding to the front moment and the rear moment is smaller than a threshold value Delta2At the moment, the torque and rotating speed data values at the moment are simultaneously rejected;
torque small load elimination threshold delta1And speed of rotation small load rejection threshold delta2The calculation formula of (2) is as follows:
△1or delta2＝(△max-△min)×θ×2 (1)
Wherein, ΔmaxIs the maximum amplitude of the torque or speed load cycle collected; deltaminIs the minimum amplitude of the torque or rotational speed load cycle collected; θ is the small load rejection threshold.
4. The method for load reproduction and extrapolation for internal combustion engine of construction machinery as claimed in claim 1, wherein the step (3) of resampling torque and speed data is as follows:
(3.1) in order to conveniently and later perform synchronous counting and improve the counting precision, the time step between two data points is as short as possible and is equally spaced, so that the torque and the rotating speed are re-sampled, the sampling rate is improved, and the time step between the two load data points before and after is shortened;
in order to increase the resampling rate to an integral multiple of the original sampling frequency in order to enable more load data points to fall on the real collected load data points during resampling, a time step delta t between two previous data points and two next data points after resampling is as follows:
△t＝1/mf (2)
wherein, f is the original sampling frequency; m is the ratio of the resampling frequency to the original sampling frequency;
(3.2) because the sampling frequency is improved, some data points are empty during resampling, and at this time, an interpolation method is adopted to complement sample points with empty data, that is, if a data point at a certain moment is empty during resampling, two points are interpolated once by using two adjacent data points at the moment, and the formula is as follows:
wherein
The moment when the data point is empty; y isiThe data point value after the linear interpolation at the moment is taken as the data point value; x is the number ofi+1The latter moment; y isi+1The data point value at the later moment; x is the number ofi-1The previous moment; y isi-1The value of the data point at the previous time.
5. The method for load reconstruction and extrapolation for internal combustion engines of construction machines according to claim 1, wherein the threshold values in step (5) are selected on the same basis for speed and torque data: the extreme sample volume produced at this threshold being no greater than the total collected sample volume
Namely:
wherein n isuIs the extreme sample capacity; n is the total collected sample volume;
the value of (c) is determined by the severity of the load change.
6. The method for regenerating and extrapolating the load of the internal combustion engine of the construction machinery as claimed in claim 1, wherein when the proportion of the low load interval in step (6) is extrapolated:
determining threshold values of the torque and the rotating speed from the step (5), when the interval of the torque and the rotating speed in the step (4) is below the respective threshold value, indicating that the load of the interval is the medium-low load, directly carrying out proportional extrapolation on the accumulated time of the interval of the medium-low load counted in the step (4), wherein a proportional extrapolation coefficient k is as follows:
k＝T×w％/t (5)
wherein T is the expected life of the internal combustion engine of the engineering machinery; w% is the ratio of the medium and low load working condition; t is the time at which the sample load is taken.
7. The method for load reconstruction and extrapolation for internal combustion engine of construction machine according to claim 1, wherein in the extreme value extrapolation of POT model in extreme value load interval of step (7):
for the accumulated time of the interval in which the extreme load is located, the proportional extrapolation cannot be directly carried out as in the step (6), considering the diversity of the extreme load, the extreme extrapolation of the POT model is firstly carried out, and then the extrapolated load time sequence is synchronously counted by torque and rotating speed, wherein the specific steps of the POT model extrapolation are as follows:
(7.1) generalized pareto distribution GPD fitting extreme load
The cumulative distribution function expression for GPD is:
wherein sigma is a scale parameter; xi is a shape parameter; u is a threshold value;
(7.2) extreme extrapolation
When the POT model is used for extrapolation, only the extreme load is extrapolated, the extreme value after extrapolation is the result of increasing the GPD probability corresponding to the original extreme load by alpha%, and the alpha value is determined by the intensity of load change under different working conditions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110620349.5A CN113190933B (en) | 2021-06-03 | 2021-06-03 | Load reproducing and extrapolation method for internal combustion engine of engineering machinery |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110620349.5A CN113190933B (en) | 2021-06-03 | 2021-06-03 | Load reproducing and extrapolation method for internal combustion engine of engineering machinery |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113190933A CN113190933A (en) | 2021-07-30 |
| CN113190933B true CN113190933B (en) | 2022-06-24 |
Family
ID=76975853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110620349.5A Expired - Fee Related CN113190933B (en) | 2021-06-03 | 2021-06-03 | Load reproducing and extrapolation method for internal combustion engine of engineering machinery |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113190933B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103674571A (en) * | 2014-01-03 | 2014-03-26 | 重庆市科学技术研究院 | Automatic transmission loading spectrum collecting and compiling method used for indoor bench test |
| CN106528987A (en) * | 2016-11-03 | 2017-03-22 | 河北工业大学 | Method for accumulated damage computation and life prediction of IGBT module used for electric car |
| DE102016110412A1 (en) * | 2015-12-11 | 2017-06-14 | Hyundai Motor Company | Method and device for active vibration control of a hybrid electric vehicle |
| CN106934184A (en) * | 2017-04-25 | 2017-07-07 | 吉林大学 | A kind of Digit Control Machine Tool load Extrapolation method based on the extension of time domain load |
| JP2018119886A (en) * | 2017-01-26 | 2018-08-02 | いすゞ自動車株式会社 | Estimation device and estimation method |
| CN112393891A (en) * | 2020-11-23 | 2021-02-23 | 中国农业大学 | Wireless monitoring system and method for fatigue damage of key parts of agricultural operation machinery |
-
2021
- 2021-06-03 CN CN202110620349.5A patent/CN113190933B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103674571A (en) * | 2014-01-03 | 2014-03-26 | 重庆市科学技术研究院 | Automatic transmission loading spectrum collecting and compiling method used for indoor bench test |
| DE102016110412A1 (en) * | 2015-12-11 | 2017-06-14 | Hyundai Motor Company | Method and device for active vibration control of a hybrid electric vehicle |
| CN106528987A (en) * | 2016-11-03 | 2017-03-22 | 河北工业大学 | Method for accumulated damage computation and life prediction of IGBT module used for electric car |
| JP2018119886A (en) * | 2017-01-26 | 2018-08-02 | いすゞ自動車株式会社 | Estimation device and estimation method |
| CN106934184A (en) * | 2017-04-25 | 2017-07-07 | 吉林大学 | A kind of Digit Control Machine Tool load Extrapolation method based on the extension of time domain load |
| CN112393891A (en) * | 2020-11-23 | 2021-02-23 | 中国农业大学 | Wireless monitoring system and method for fatigue damage of key parts of agricultural operation machinery |
Non-Patent Citations (2)
| Title |
|---|
| A Review of the Extrapolation Method;Jixin Wang;《Strojniški vestnik - Journal of Mechanical Engineering》;20161230;全文 * |
| 基于极值外推的高速列车齿轮;宫海彬等;《吉林大学学报》;20140930;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113190933A (en) | 2021-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106886638B (en) | A kind of endless-track vehicle transmission shaft loading spectrum preparation method based on Density Estimator | |
| CN109213654A (en) | A kind of method for detecting abnormality and device | |
| CN111089726A (en) | Rolling bearing fault diagnosis method based on optimal dimension singular spectrum decomposition | |
| CN104750830B (en) | The cycle method for digging of time series data | |
| CN113538353B (en) | Five-phase asynchronous motor rolling bearing fault diagnosis method based on single-channel diagram data enhancement and migration training residual error network | |
| CN110942137A (en) | Power grid information operation and maintenance monitoring method based on deep learning | |
| CN113190933B (en) | Load reproducing and extrapolation method for internal combustion engine of engineering machinery | |
| CN112883075B (en) | Landslide universal type ground surface displacement monitoring data missing and outlier processing method | |
| CN102998118A (en) | Bearing quantitative diagnosis method based on morphological filtering and complexity measure | |
| CN116481821A (en) | Engine fault early warning method and system based on big data analysis management | |
| CN109682597A (en) | A kind of gear-box vibration signal processing and analysis method | |
| CN115100819B (en) | Landslide hazard early warning method and device based on big data analysis and electronic equipment | |
| CN115879245A (en) | Method for evaluating long-life transmission fatigue reliability of medium carbon steel gear | |
| CN111814110A (en) | Bridge health monitoring data control chart analysis method | |
| CN114969645A (en) | Statistical distribution model-based aircraft engine state monitoring and evaluation method | |
| CN111581603A (en) | Load spectrum processing method and device | |
| CN112621381B (en) | Intelligent health state evaluation method and device for machine tool feeding system | |
| CN102592172A (en) | Planet wheel fault recognition method based on sensitive measure point data fusion | |
| CN105373583A (en) | Modeling method for support vector machine based on data compression | |
| CN114112390B (en) | Nonlinear complex system early fault diagnosis method | |
| CN115421400A (en) | Digital twin virtual-real consistency discrimination method for service health state intelligent recognition decision | |
| CN115030903A (en) | On-line diagnosis method for early fault of rolling bearing in centrifugal pump | |
| CN112698405A (en) | Landslide event identification, early warning and evaluation method based on seismic signal | |
| CN114735015B (en) | Load assessment method for commercial diesel vehicle | |
| CN112485748A (en) | Phase-to-phase judgment method for single-phase electric meter |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220624 |