CN115164812B - Angle measuring method of EPS system angle sensor based on sensor model - Google Patents
Angle measuring method of EPS system angle sensor based on sensor model Download PDFInfo
- Publication number
- CN115164812B CN115164812B CN202210905002.XA CN202210905002A CN115164812B CN 115164812 B CN115164812 B CN 115164812B CN 202210905002 A CN202210905002 A CN 202210905002A CN 115164812 B CN115164812 B CN 115164812B
- Authority
- CN
- China
- Prior art keywords
- signal
- angle
- sensor
- pwm
- sensor model
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims description 21
- 238000005259 measurement Methods 0.000 claims abstract description 35
- 238000000691 measurement method Methods 0.000 claims abstract description 12
- 210000002569 neuron Anatomy 0.000 claims abstract description 8
- 238000004364 calculation method Methods 0.000 claims description 30
- 230000006870 function Effects 0.000 claims description 17
- 239000011159 matrix material Substances 0.000 claims description 15
- 230000014509 gene expression Effects 0.000 claims description 14
- 230000004913 activation Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 description 7
- 238000006243 chemical reaction Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 239000002131 composite material Substances 0.000 description 4
- 238000012549 training Methods 0.000 description 3
- 239000008358 core component Substances 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 239000000306 component Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
- G01B21/22—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring angles or tapers; for testing the alignment of axes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C1/00—Measuring angles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M17/00—Testing of vehicles
- G01M17/007—Wheeled or endless-tracked vehicles
- G01M17/06—Steering behaviour; Rolling behaviour
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Length Measuring Devices With Unspecified Measuring Means (AREA)
Abstract
The invention relates to the technical field of automobile steering systems, and provides an EPS system angle sensor angle measurement method based on a sensor model, which comprises the following steps: step 1, an ECU samples an angle sensor to obtain an angle signal of the angle sensor; step 2, converting the angle signal into a comprehensive signal with a continuity value between 0.5 and 42.5; step 3, calculating the measurement angle of the angle sensor based on the sensor model; and 4, calculating based on the reliability of the integrated signal. The sensor model in the step 3 is a three-layer network model with unitary input and unitary output, programming is easy to implement in a modularized mode, mistakes are not easy to occur, and the mode is simple and easy to derive by corresponding the combined cycle period of 43 neurons in the middle layer and signals, so that the weight configuration can be completed without data learning, the regularity is high, and the parameters of the sensor model are easy to memorize. The invention can be embedded into the ECU program in a modularized way for operation, and can improve the operation efficiency.
Description
Technical Field
The invention relates to the technical field of automobile steering systems, in particular to an angle measuring method of an EPS system angle sensor based on a sensor model.
Background
The electric power steering system (EPS for short) of the automobile is applied to a steering mechanism of the automobile, can enable a driver to steer lightly, can timely and accurately execute steering instructions, and assists the driver in steering operation. The angle sensor is used as one of core components of the EPS system, whether the angle measurement is accurate and reliable or not, the performance of the EPS system is directly affected, the angle signal is indispensable for realizing functions such as active alignment, even chassis electric control systems such as ESP and the like need to share the angle signal with the EPS to complete respective control functions, and therefore integrated control of the vehicle chassis is realized. The angle sensor of the EPS system adopts non-contact type, the output angle signal of the angle sensor also adopts a duty cycle PWM mode, an electronic control unit (ECU for short) in the EPS system is required to sample and identify the angle sensor, and the angle (also called as a measured angle) is calculated through a proper algorithm.
At present, the methods commonly adopted in the industry are: the angle measurement is completed by embedding a vernier algorithm program in the ECU, and the algorithm adopts more calculation processes, multi-signal combination calculation, logic judgment, table lookup search and the like, so that modularization cannot be formed, programming and implementation are required one by one, the workload is large, errors are easy to occur, the IC resources of the ECU are occupied, and the calculation efficiency is also influenced; meanwhile, in order to adapt to the current situation of the ECU chip at the moment, the common vernier algorithm adopts a 16-bit binary maximum number 65535 rounding conversion algorithm, and more similar rounding conversion is included in the whole process of complete calculation, so that a part of calculation accuracy is lost, the grade of the ECU chip is improved, the vernier algorithm is still executed according to the 16-bit binary numerical calculation accuracy, and a measurement accuracy improvement space is not provided.
Disclosure of Invention
The angle measurement method mainly solves the technical problems that angle measurement is completed through a vernier algorithm in the prior art, modularization cannot be formed, programming and implementation are needed one by one, workload is large, errors are easy to occur, IC resources of an ECU are occupied, the operation efficiency is influenced and the like, and therefore the angle measurement method for the EPS system angle sensor based on the sensor model is provided. According to the P signal and the S signal output by the angle sensor obtained by ECU sampling, the P signal and the S signal are converted into a comprehensive signal delta PWM, then a sensor model is utilized to identify the P signal section number Kp, and the comprehensive signal is utilized to directly calculate the error number Ks of the S signal, so that angle measurement and reliability evaluation are realized. The method is embedded into an ECU program to run to complete the angle measurement function.
The invention provides an EPS system angle sensor angle measurement method based on a sensor model, which comprises the following steps:
Step 1, an ECU samples an angle sensor to obtain an angle signal of the angle sensor, wherein the angle signal comprises a P signal and an S signal; wherein the P signal is denoted PWM_P and the S signal is denoted PWM_S;
step 2, converting the angle signal into a comprehensive signal delta PWM with a continuity value between 0.5 and 42.5 by using the following formula;
ΔPWM ＝ 4×(37×PWM_S-5×PWM_P-4)/3+5.5 (1)
Wherein, the P signal and the S signal can be combined at will between 0.125 and 0.875 respectively, and the delta PWM numerical interval is between 0.5 and 42.5;
Step 3, calculating a measurement angle of the angle sensor based on the sensor model, including the following steps 301 to 304:
Step 301, establishing a basic structure of a sensor model;
Step 302, determining the weight of a sensor model on the basis of the basic structure of the sensor model to obtain the sensor model;
step 303, obtaining a P signal section number Kp through the obtained sensor model;
step 304, after obtaining the P signal segment number Kp through the sensor model, combining with pwm_p, and calculating the measurement angle θ using the following formula:
θ＝40×(Kp-1)-740+40×(PWM_P-0.125)/0.75 (7)。
Further, in step 301, a basic structure of the sensor model is established as follows:
The input layer code of the sensor model is x i,i＝0、1,x0 =1, which is the bias input of the model, x 1 is the integrated signal input, and x 1 =Δpwm;
The code number of the middle layer is Z k, 43 neurons are total, k=1, 2, … and 43, the activation functions of the middle layer are all step functions, the code number is f, and the output code number of each neuron after being processed by the activation functions is Z k;
The code number of the output layer is Y, and the Y value of the output layer outputs a P signal section number Kp after linear processing;
The weight code of the input layer and the middle layer is V k,i, and the weight code of the middle layer and the output layer is W k;
The feedforward calculation expression of the sensor model is as follows:
further, the equivalent expressions of the formulas (2) and (4) which can be converted into the matrix form are as follows:
the equivalent expression of the matrix mode of the formula (2) is:
the equivalent expression of the matrix mode of the formula (4) is:
further, in step 302, weights V k,i and W k are given as follows:
[Wk]1×43＝[37,-22,15,-22,15,-22,15,15,-22,15,-22,15,...,15,-22,15,-22,15,-22]；
wherein the weight matrix [ W k]1×43 ] has 43 elements in total, and the rest 40 elements except the first 2 elements and the last 1 element are repeated by 5 elements as a group, and 8 groups are all provided.
Further, after step 3, the method further includes:
and 4, calculating based on the reliability of the integrated signal.
Further, step 4, the reliability calculation based on the integrated signal includes the following steps:
According to the integrated signal delta PWM, calculating an error number Ks of the S signal, wherein the calculation formula is as follows:
Ks＝ROUNDDOWN(ΔPWM) (8)
wherein ROUNDDOWN () is a round down function;
According to the error number Ks of the S signal, the formula for calculating the reliability Rel is as follows:
Rel＝1-|2ΔPWM-2Ks-1| (9)。
Compared with the prior art, the angle measurement method of the EPS system angle sensor based on the sensor model has the following advantages:
1. According to the combination rule of the P signal and the S signal of the angle sensor of the EPS system, if the P signal and the S signal are combined after being cross multiplied according to the cycle period of the P signal and the S signal, the formed comprehensive signal delta PWM contains the relevant characteristics and information of the combination of the P signal and the S signal and can be used for calculating the angle theta to finish the angle measurement function of the EPS system; the invention converts the P signal and the S signal of the angle sensor into the comprehensive signal delta PWM, and then utilizes the sensor model to realize angle measurement, thereby realizing domestic application and popularization of the EPS system sensor and being beneficial to cost control.
2. The invention can centralize the calculation process into unitary input and unitary output by establishing the sensor model, and the P signal section number Kp is calculated by the comprehensive signal delta PWM directly through the model, so that the programming is easy to implement in a modularized manner, the error is not easy to occur, and more calculation processes in the conventional algorithm and the combination calculation and logic judgment of multiple signals are avoided.
3. The sensor model of the invention corresponds to the combination rule of the P signal and the S signal of the angle sensor corresponding to the comprehensive signal delta PWM by using the combination cycle period of 43 neurons in the middle layer and the signals, so that the model is simple and easy to deduce, the configuration of the weight can be completed without data learning, the regularity of the weight matrix is strong, and the weight parameters of the sensor model are easy to memorize.
4. The method adopts the sensor model to implement angle calculation, replaces the modes of logic conversion, rounding calculation, table lookup search and the like of the conventional algorithm in the industry, can be modularly embedded into the ECU program for operation, has fewer occupied IC resources, can improve the operation efficiency, can adapt to higher-level ECU chip application, and has larger measurement precision improving space.
Drawings
FIG. 1 is a graph showing the characteristics of an angle signal output by an angle sensor in an EPS system when an S signal applied by the invention has no deviation;
FIG. 2 is a graph showing the combination relationship between the P signal and the S signal of the EPS system angle sensor when the S signal converted by the invention has no deviation;
FIG. 3 is a graph of the relationship between the integrated signal ΔPWM and the P/S signal in the measurement angle when the S signal provided by the invention has no deviation;
FIG. 4 is a graph showing the distribution of P signal number Kp on integrated signal ΔPWM provided by the present invention;
FIG. 5 is a graph of reliability Rel versus integrated signal ΔPWM provided by the present invention;
FIG. 6 is a flowchart of an implementation of the angle measurement method of the EPS system angle sensor based on the sensor model provided by the invention;
fig. 7 is a diagram of a sensor model for calculating P signal number Kp according to the present invention.
Detailed Description
In order to make the technical problems solved by the invention, the technical scheme adopted and the technical effects achieved clearer, the invention is further described in detail below with reference to the accompanying drawings and the embodiments. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the matters related to the present invention are shown in the accompanying drawings.
The angle measuring method of the EPS system angle sensor based on the sensor model, provided by the embodiment of the invention, is applied to the vehicle EPS system, and the angle sensor is one of core components of the EPS system. As shown in fig. 1, the angle signal output by the angle sensor has a P signal and an S signal, which are output by the built-in IC chip thereof, the P signal and the S signal take the form of duty cycle, the P signal is denoted as pwm_p, and the S signal is denoted as pwm_s. The pwm_p signal adopted by the current EPS system is 40 ° one cycle, 37 cycle sections in total, and Kp (kp=1, 2, … …,36, 37) is set according to the P signal section number from left to right as shown in fig. 1; the pwm_s signal used in the current EPS system is 296 ° one cycle, 5 cycle segments in total, and the S signal segment numbers from left to right as shown in fig. 1 are set to K (k=1, 2, 3,4, 5).
The PWM_P and PWM_S variation amounts in each cycle section are in linear relation with the angle variation amount, the duty ratio value range of the effective pulse width modulation is 0.125-0.875 (12.5% -87.5%), and the angle measurement range is +/-740 degrees. The characteristic curve of the angle signal shown in fig. 1 is developed rightward by taking the lowest angle point-740 ° as a base point, and the signal value of the lowest angle point is 0.125, that is, pwm_p=pwm_s=0.125, which can be understood as a curve within any angle range of ±740°, and if the curves are connected end to end, the curves are circularly repeated. In any range of +/-740 degrees, if the angle scales of-740 degrees to +740 degrees are divided, unique PWM_P and PWM_S combinations corresponding to the unique PWM_P and PWM_S combinations exist at any angle point, namely the measurement angle theta=F (PWM_P and PWM_S) can be expressed, and if the angle exceeds the range of +/-740 degrees, repeated signal combinations can occur. Therefore, the angle measured by the present invention is a relative angle, which is effective in the range of ±740°.
In order to analyze the combination characteristics of pwm_p and pwm_s, a graph is drawn as shown in fig. 2, wherein the right and upper P signal section numbers Kp represent the curve section where the P signal section numbers Kp and S signal section numbers are combined, and the S signal difference between adjacent curve sections is 0.02027 (when the S signal is in the vicinity of the start point or end point of each cycle section of the S signal, although the S signal value is suddenly changed, the difference of the scale is unchanged, the S signal difference is considered to be 0.02027), so that the S signal has ± 0.010135 deviation, the P signal section numbers Kp corresponding to the curve sections are unique, and the P signal section numbers Kp corresponding to the curve sections are also unique, respectively, in any combination between 0.125 and 0.875.
As shown in fig. 2, when the S signal is unbiased, the P signal is clearly visible for 41 total combined curve segments of the S signal. When the S signal deviates, if the deviation changes to the direction of reducing the S signal value, the S signal curves of 5 sections in fig. 1 move to the right at the same time, the curve of the S signal k=5 sections is decomposed into 2 parts, and the upper part of the curve is pushed to the upper left in fig. 1; similarly, if the deviation is changed in the direction of increasing the S signal value, which corresponds to the 5-segment S signal curve of fig. 1 moving leftward at the same time, the curve of the S signal k=1 segment is also decomposed into 2 parts, and the lower part thereof is shifted to the far right in fig. 1. Thus, when the S signal has 2 direction deviations, and each direction is a numerical point set of the S signal deviations, the combined curve segments of the 2P signals and the S signal are increased, and if they are converted into a combined relation diagram (not shown in the drawing of this patent) containing the S signal deviations as shown in fig. 2, the combined curve segments should have 43 total.
In fig. 2, although a combined curve segment is shown with no deviation of the S signal, it can be extended to the case of a combined curve segment with deviation of the S signal. Therefore, for any pwm_p value point, 43 pwm_s value point sets appear in total, corresponding to the pwm_p value point, which is the error number Ks of the S signal.
After the curve shown in fig. 2 is analyzed and the comprehensive analysis of the curve shown in fig. 1 is combined, 37 loops of the P signal and 5 loops of the S signal are found, the P signal and the S signal are combined after being multiplied by each other to form a comprehensive signal delta PWM, the comprehensive signal delta PWM contains relevant characteristics and information of the combination of the P signal and the S signal, and a relation diagram of the comprehensive signal delta PWM and the P/S signal in a measuring angle when the S signal has no deviation as shown in fig. 3 can be drawn.
When S signal has deviation within +/-0.010135 (which is scale deviation), the integrated signal delta PWM shows the continuity of signal values, the characteristic of the continuity represents the influence degree of the S signal deviation on the effectiveness of the angle measurement, the degree is defined as credibility, the code is Rel, and the value range is 0-1 (0% -100%). Meanwhile, in the S signal deviation interval, the comprehensive signal delta PWM with the continuity value change can still correspond to the uniqueness of the P signal number Kp accurately; the error number Ks of the S signal corresponds to all the numerical points of the PWM_S within the deviation interval of + -0.010135, and the PWM_S deviation changesAlways with ΔPWM over two valuesWhich correspond linearly. The delta PWM expression is as follows, after being arranged, and the value of the continuity of the integrated signal is in a relatively easily identifiable interval:
ΔPWM ＝ 4×(37×PWM_S-5×PWM_P-4)/3+5.5 (1)
wherein, the P signal and the S signal can be combined at will between 0.125 and 0.875, and the delta PWM value interval is between 0.5 and 42.5.
The following table 1 can be obtained from the relationship shown in the formula (1) and fig. 2 and 3.
Table 1: corresponding data table of integrated signal delta PWM and P signal number Kp and credibility Rel
According to table 1, a distribution diagram of the P signal section number Kp on the integrated signal Δpwm as shown in fig. 4 can be obtained, and a relationship diagram of the reliability Rel and the integrated signal Δpwm as shown in fig. 5 can be obtained.
After the P signal and the S signal of the angle sensor are converted into the integrated signal Δpwm, the data in table 1 and the curve rules shown in fig. 4 and 5 are analyzed, and the flow of the angle measurement method of the EPS system angle sensor based on the sensor model provided by the embodiment of the invention, as shown in fig. 6, includes the following procedures:
And step 1, the ECU samples an angle sensor to acquire an angle signal of the angle sensor, wherein the angle signal comprises a P signal and an S signal. Wherein the P signal is denoted pwm_p and the S signal is denoted pwm_s.
Before the EPS system is installed, the angle sensor is calibrated by equipment, and when the S signal is in an ideal state without deviation, two angle signal curves output by the built-in IC chip are shown in figure 1. Since the factors influencing the P signal are fewer, the deviation can be defaulted to be generated, and the factors influencing the S signal are more, but when the angle sensor product is discharged, the deviation of the S signal can be always controlled within the range of +/-0.010135. Therefore, when the S signal deviates within the range of +/-0.010135, the ECU samples the acquired P signal and the S signal, which are respectively in any combination between 0.125 and 0.875.
And 2, converting the angle signal into a comprehensive signal with a continuity value between 0.5 and 42.5.
The angle signal is converted into a composite signal according to the following calculation formula of the composite signal delta PWM:
ΔPWM ＝ 4×(37×PWM_S-5×PWM_P-4)/3+5.5 (1)
and step 3, calculating the measurement angle of the angle sensor based on the sensor model.
The sensor is a neural network model, and can rapidly and reliably solve the problem of linear separability. The sensor model has learning capability, correct weight configuration parameters can be obtained through learning of training data, and the feedforward calculation is performed on the sensor model built according to the correct weight configuration parameters, so that the corresponding input can obtain correct output.
According to the data of table 1 and the analysis of the curve of fig. 4, the sensor model can be utilized to calculate the P signal section number Kp by the integrated signal Δpwm, so that the calculation process is concentrated into unitary input and unitary output, the modular programming is easy to realize, the error is not easy, and the more calculation processes in the vernier algorithm and the combination of table lookup are avoided.
Step 301, a basic structure of a sensor model is established.
Building a basic structure of a sensor model as shown in fig. 7, wherein the code of an input layer of the sensor model is x i,i＝0、1,x0 =1, which is the offset input of the model, and x 1 is the integrated signal input, namely x 1 =Δpwm; the code number of the middle layer is Z k, 43 neurons are total, k=1, 2, … and 43, the activation functions of the middle layer are all step functions, the code number is f, and the output code number of each neuron after being processed by the activation functions is Z k; the code number of the output layer is Y, and the Y value is linearly processed to output the P signal section number Kp. The weight code of the input layer and the middle layer is V k,i, and the weight code of the middle layer and the output layer is W k. The feedforward calculation expression of the sensor model is as follows:
To facilitate the implementation of the model by programming, the equivalent expressions of equations (2) and (4) can be converted into matrix form as follows.
The equivalent expression of the matrix mode of the formula (2) is:
the equivalent expression of the matrix mode of the formula (4) is:
step 302, determining the weight of the sensor model based on the basic structure of the sensor model to obtain the sensor model.
The model belongs to a simple and easy-to-deduce model, the configuration of the data learning completion weight can be omitted, and the weight of the model can be obtained through learning training.
In the case of no data learning, the weights V k,i and W k given by the present invention are as follows:
[Wk]1×43＝[37,-22,15,-22,15,-22,15,15,-22,15,-22,15,...,15,-22,15,-22,15,-22]；
Wherein the total of 43 elements in the weight matrix [ W k]1×43 ] and the rest of 40 elements except the first 2 elements and the last 1 elements are repeated by taking 5 elements as a group, and the total of 8 groups is that the weight matrix [ W k]1×43 ] is that '…' represents the rest of 5 '15, -22, 15, -22, 15' element groups.
The sensor model has data learning capability, and under the condition of learning training, a weight parameter matrix is configured through the learning of the model. When different initialization parameters and learning rates are set, the model is configured with countless weight parameter matrixes, and the weight parameter matrixes can meet the requirement of calculating the P signal section number Kp within a specified precision range.
For example: in the case where the weight parameter [ W k]1×43 ] between the intermediate layer and the output layer is unchanged, the configuration parameter of one of the weights [ V k,i]43×2 ] between the input layer and the intermediate layer of the model is as follows:
Step 303, obtaining the P signal segment number Kp through the obtained sensor model.
Step 304, after obtaining the P signal segment number Kp through the sensor model, combining with pwm_p to calculate the measurement angle θ, where the expression is:
θ＝40(Kp-1)-740+40(PWM_P-0.125)/0.75 (7)
step 4, reliability calculation based on the integrated signal
Firstly, calculating an error number Ks of an S signal according to a comprehensive signal delta PWM, wherein the calculation formula is as follows:
Ks＝ROUNDDOWN(ΔPWM) (8)
In the formula, ROUNDDOWN () is a downward rounding function.
According to the error number Ks of the S signal, the formula for calculating the reliability Rel is as follows:
Rel＝1-|2ΔPWM-2Ks-1| (9)。
In the numerical range of 0.125-0.875 of the angle signal, retaining the effective numerical value of 4 bits after decimal points of the angle signal, randomly extracting 100 PWM_P and PWM_S combinations, and comparing and verifying the angle measurement method with a common vernier algorithm by utilizing the angle measurement method, wherein the angle measurement value and the credibility value are shown in the following table:
And (3) carrying out multiple verification according to the comparison verification method, wherein the difference value of the angle measurement values obtained by the two algorithms is smaller than I+/-0.003I DEG, and the difference value of the obtained credibility values is smaller than I+/-0.3 I%. The measurement accuracy of the invention meets the angle measurement requirement of the EPS system.
It should be noted that: in order to adapt to the current situation of the ECU chip at the time, a common vernier algorithm adopts a 16-bit binary maximum number 65535 rounding conversion algorithm, the corresponding range of the vernier algorithm and the duty ratio of 0-100% is 0-65535, and the corresponding formula is as follows: pwm_p_pc=pwm_p 65535 and pwm_s_pc=pwm_s 65535, then performing rounding calculation, and in the whole process of the complete calculation, more similar rounding conversion is included, so that a part of calculation accuracy is lost, even if the level of the ECU chip is further improved, the vernier algorithm is still executed according to 16-bit binary numerical calculation accuracy, the accuracy still can meet the angle measurement requirement of the EPS system, but for integrated control of the vehicle chassis, the measurement angle and the reliability are borrowed by other functional components such as ESP (electronic body stability control system), and the measurement accuracy at the moment is not enough to support the systems; the angle measurement of the invention is based on a 32-bit ECU chip commonly adopted in the industry at present, IQ 19-IQ 21 (representing a 32-bit binary floating point numerical value different precision calculation module) can be called according to the angle measurement, the calculation can be performed within the precision of 0.000001907-0.000000477, and the calculation result is more accurate. For example: the comparative measurements of three typical angle signal points are shown in the following table, the angle measurement and the confidence value of the three points being theoretically known and being accurate values without errors.
When the ECU adopts higher level chip, the measurement accuracy of this patent still has great promotion space.
In summary, the invention provides a sensor model-based angle measurement method for an EPS system angle sensor, which is implemented by sampling a P signal and an S signal output by the angle sensor by an ECU, converting the P signal and the S signal into a composite signal Δpwm, identifying the number Kp of the P signal by using the sensor model, and directly calculating the error number Ks of the S signal by using the composite signal. The method is embedded into an ECU program to run to complete the angle measurement function.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments is modified or some or all of the technical features are replaced equivalently, so that the essence of the corresponding technical scheme does not deviate from the scope of the technical scheme of the embodiments of the present invention.
Claims (5)
1. The angle measuring method of the EPS system angle sensor based on the sensor model is characterized by comprising the following steps of:
Step 1, an ECU samples an angle sensor to obtain an angle signal of the angle sensor, wherein the angle signal comprises a P signal and an S signal; wherein the P signal is denoted PWM_P and the S signal is denoted PWM_S;
step 2, converting the angle signal into a comprehensive signal delta PWM with a continuity value between 0.5 and 42.5 by using the following formula;
ΔPWM ＝ 4×(37×PWM_S-5×PWM_P-4)/3+5.5 (1)
Wherein, the P signal and the S signal can be combined at will between 0.125 and 0.875 respectively, and the delta PWM numerical interval is between 0.5 and 42.5;
Step 3, calculating a measurement angle of the angle sensor based on the sensor model, including the following steps 301 to 304:
Step 301, a basic structure of a sensor model is established as follows:
The input layer code of the sensor model is x i,i＝0、1,x0 =1, which is the bias input of the model, x 1 is the integrated signal input, and x 1 =Δpwm;
The code number of the middle layer is Z k, 43 neurons are total, k=1, 2, … and 43, the activation functions of the middle layer are all step functions, the code number is f, and the output code number of each neuron after being processed by the activation functions is Z k;
The code number of the output layer is Y, and the Y value of the output layer outputs a P signal section number Kp after linear processing;
The weight code of the input layer and the middle layer is V k,i, and the weight code of the middle layer and the output layer is W k;
The feedforward calculation expression of the sensor model is as follows:
Step 302, determining the weight of a sensor model on the basis of the basic structure of the sensor model to obtain the sensor model;
step 303, obtaining a P signal section number Kp through the obtained sensor model;
step 304, after obtaining the P signal segment number Kp through the sensor model, combining with pwm_p, and calculating the measurement angle θ using the following formula:
θ＝40×(Kp-1)-740+40×(PWM_P-0.125)/0.75 (7)。
2. The angle measurement method of an EPS system angle sensor based on a sensor model according to claim 1, wherein the equivalent expressions of the formulas (2) and (4) convertible into a matrix form are as follows:
the equivalent expression of the matrix mode of the formula (2) is:
the equivalent expression of the matrix mode of the formula (4) is:
3. The method for angle measurement of an EPS system angle sensor based on a sensor model as set forth in claim 2, wherein in step 302, weights V k,i and W k are given as follows:
[Wk]1×43＝[37,-22,15,-22,15,-22,15,15,-22,15,-22,15,...,15,-22,15,-22,15,-22]；
wherein the weight matrix [ W k]1×43 ] has 43 elements in total, and the rest 40 elements except the first 2 elements and the last 1 element are repeated by 5 elements as a group, and 8 groups are all provided.
4. A method for angle measurement of an EPS system angle sensor based on a sensor model as set forth in claim 1 or 3, further comprising, after step 3:
and 4, calculating based on the reliability of the integrated signal.
5. The angle measurement method of an EPS system angle sensor based on a sensor model as set forth in claim 4, wherein the step 4 of calculating the reliability based on the integrated signal includes the steps of:
According to the integrated signal delta PWM, calculating an error number Ks of the S signal, wherein the calculation formula is as follows:
Ks＝ROUNDDOWN(ΔPWM) (8)
wherein ROUNDDOWN () is a round down function;
According to the error number Ks of the S signal, the formula for calculating the reliability Rel is as follows:
Rel＝1-|2ΔPWM-2Ks-1| (9)。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210905002.XA CN115164812B (en) | 2022-07-29 | 2022-07-29 | Angle measuring method of EPS system angle sensor based on sensor model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210905002.XA CN115164812B (en) | 2022-07-29 | 2022-07-29 | Angle measuring method of EPS system angle sensor based on sensor model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115164812A CN115164812A (en) | 2022-10-11 |
| CN115164812B true CN115164812B (en) | 2024-08-02 |
Family
ID=83478066
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210905002.XA Active CN115164812B (en) | 2022-07-29 | 2022-07-29 | Angle measuring method of EPS system angle sensor based on sensor model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115164812B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105466332A (en) * | 2015-11-13 | 2016-04-06 | 珠海格力节能环保制冷技术研究中心有限公司 | Angle sensor and angle measuring method |
| CN110712201A (en) * | 2019-09-20 | 2020-01-21 | 同济大学 | Robot multi-joint self-adaptive compensation method based on perceptron model and stabilizer |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE10257559B4 (en) * | 2002-12-10 | 2013-02-07 | Robert Bosch Gmbh | Determination of the start segment of a rotation angle sensor with periodic characteristic |
| CN113063386B (en) * | 2021-03-22 | 2022-09-30 | 大连创新零部件制造公司 | Angle measuring method of EPS system angle sensor |
-
2022
- 2022-07-29 CN CN202210905002.XA patent/CN115164812B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105466332A (en) * | 2015-11-13 | 2016-04-06 | 珠海格力节能环保制冷技术研究中心有限公司 | Angle sensor and angle measuring method |
| CN110712201A (en) * | 2019-09-20 | 2020-01-21 | 同济大学 | Robot multi-joint self-adaptive compensation method based on perceptron model and stabilizer |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115164812A (en) | 2022-10-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106698642A (en) | Multi-objective real-time optimization control method for sewage treatment process | |
| CN113537319B (en) | Wind tunnel balance lifelong learning and calibrating method and system for typed model | |
| CN102365538A (en) | Bearings for wheels with sensors | |
| CN109684677A (en) | A kind of gradient evaluation method based on Kalman filtering algorithm | |
| CN104660260B (en) | The adaptive-filtering digital calibration circuit and method of ADC | |
| CN115164812B (en) | Angle measuring method of EPS system angle sensor based on sensor model | |
| CN115755589B (en) | An optimization method for reactor fuzzy control system based on genetic algorithm | |
| CN118761158A (en) | Cascade estimation method and system for road adhesion coefficient considering model parameter mismatch | |
| CN113139336A (en) | Water quality system parameter identification method based on fuzzy fractional order neural network | |
| Holzmann et al. | Vehicle dynamics simulation based on hybrid modeling | |
| CN113063386B (en) | Angle measuring method of EPS system angle sensor | |
| Meng et al. | A homogeneous domination control method based on sampled‐data output feedback for lateral stability of an intelligent electric vehicle | |
| CN111062088A (en) | Four-wheel independent drive electric vehicle reference speed fault-tolerant self-adaptive estimation method | |
| Kim et al. | Diagnosis-based design of electric power steering system considering multiple degradations: role of designable generative adversarial network anomaly detection | |
| CN109885876A (en) | A real-time modeling method of sliding mode observer for fault diagnosis of electric power steering system | |
| CN119416161A (en) | A method and device for detecting steering performance of mining vehicles | |
| Lee et al. | Finite-memory front-wheel angle estimation strategy for steer-by-wire system | |
| Sinz et al. | SAT-based consistency checking of automotive electronic product data | |
| CN118153219A (en) | A design method for variable transmission ratio of steer-by-wire based on Gaussian regression | |
| CN113515807B (en) | Wheel vertical force detection method and device, electronic equipment and storage medium | |
| CN118863588B (en) | A quantitative evaluation method for tire and vehicle matching | |
| CN116048001A (en) | A Quasi-static Key Error Identification Method of CNC Gear Grinding Machine Based on Tooth Pitch Error | |
| CN111055920B (en) | A Construction Method of Multi-Model Corner Controller for Automotive EPS Steering System | |
| Meng et al. | A Self-Organizing Modular Neural Network for Nonlinear System Modeling | |
| Im et al. | Design of steer-by-wire system with bilateral control method using disturbance observer |
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 |