Method for estimating intake pressure of hydrogen engine

Abstract

The invention provides a hydrogen engine intake pressure estimation method, which comprises the following steps of: dividing working modes, namely dividing the working modes into a small load working condition, a medium load working condition and a large load working condition according to the opening degree of a throttle valve of a hydrogen engine; the hydrogen engine is operated under different working conditions, the actual value of the air inlet pressure and the estimated value of each observer are collected in real time every T moment, and the numerical value of each evaluation index is calculated; carrying out numerical standardization and normalization processing on the evaluation indexes; determining the weight of each evaluation index through an entropy weight method; calculating a comprehensive score of the observer through linear weighting; and selecting the observer with the highest comprehensive score as the optimal observer according to the comprehensive scores of the observers, and selecting the estimated value of the optimal observer as the air inlet pressure output of the hydrogen engine. The problem that the air inlet pressure cannot be measured due to the fact that the pressure sensor is easily damaged due to tempering of the air inlet hydrogen engine can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrogen engine control, and in particular to a method for estimating the intake pressure of a hydrogen engine. Background Art

[0002] As an efficient and clean power plant, hydrogen engines have broad application prospects in areas such as automobiles. Intake pressure is a key parameter for hydrogen engine combustion control. Accurate intake pressure is crucial for controlling the engine's air-fuel ratio, optimizing power output, and improving emissions performance. However, hydrogen engines face complex operating conditions during operation. The intake system is affected by various factors, including temperature, airflow pulsation, and sensor noise. In particular, backfire in the intake manifold of hydrogen engines can easily damage the pressure sensor, making direct intake pressure measurement challenging. Therefore, estimating intake pressure through an observer is of great significance.

[0003] Currently, observers used for intake pressure estimation mainly include the extended Kalman filter observer (EKF), the unscented Kalman filter observer (UKF), and the sliding mode observer (SMO). However, the performance of a single observer varies under different operating conditions, making it difficult to achieve high-precision intake pressure estimation across the entire operating range, from light load to medium load to high load.

[0004] The dynamic characteristics of intake pressure vary significantly under different operating conditions. For example, under low-load conditions, the intake pressure signal has a small amplitude and changes gently, making it susceptible to noise interference. This requires the observer to have excellent noise suppression capabilities and robust stability. On the other hand, under high-load conditions, the intake pressure changes dramatically and at a high rate, requiring a fast response capability from the observer. A single observer cannot maintain high-precision estimates under all conditions.

[0005] Therefore, how to select the optimal observer for intake pressure estimation according to different working conditions becomes a key issue to improve estimation accuracy and reliability. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, the present invention provides a method for estimating the intake pressure of a hydrogen engine. By dividing the engine operating conditions, establishing a comprehensive evaluation index system for observers under various operating conditions, and determining the optimal observer selection strategy, accurate estimation of the intake pressure under different operating conditions can be achieved. This can solve the problem of the inability to measure the intake pressure of a hydrogen engine due to the pressure sensor being easily damaged due to backfire in the intake duct.

[0007] The present invention achieves the above technical objectives through the following technical means.

[0008] A method for estimating the intake pressure of a hydrogen engine comprises the following steps:

[0009] Working mode classification: According to the throttle opening of the hydrogen engine, the working mode is divided into light load condition, medium load condition and heavy load condition;

[0010] Run the hydrogen engine under different working conditions, collect the actual value of the intake pressure and the estimated value of each observer at every T time, and calculate the value of each evaluation index, which is recorded as b ij , b ij represents the value of the jth evaluation indicator of the i-th observer; the values ​​of the evaluation indicators include mean absolute error, maximum absolute error, mean relative error, determination coefficient, maximum convergence time, average convergence time, maximum jitter amplitude and average jitter amplitude;

[0011] Standardization and normalization of the numerical values ​​of evaluation indicators;

[0012] The weight of each evaluation indicator is determined by the entropy weight method;

[0013] The comprehensive score of the observer is calculated by linear weighting;

[0014] According to the comprehensive scores of each observer, the observer with the highest comprehensive score is selected as the optimal observer, and its estimated value is selected as the intake pressure output of the hydrogen engine.

[0015] Furthermore, the operating mode is divided into light load condition, medium load condition and heavy load condition according to the throttle opening of the hydrogen engine. The specific classification criteria are as follows:

[0016] When the throttle opening satisfies θ≤θ small.max When , the hydrogen engine is in a low-load condition;

[0017] When the throttle opening satisfies θ small.max <θ≤θ med.max When , the hydrogen engine is in medium load condition;

[0018] When the throttle opening satisfies θ med.max When <θ≤100%, the hydrogen engine is in a high-load condition;

[0019] Where: θ represents the throttle opening; θ small.max The maximum throttle opening allowed under low load conditions; θ med.max It is the maximum throttle opening allowed under medium load conditions.

[0020] Furthermore, the value of the j evaluation index of the i-th observer is recorded as b ij , j∈[1,2,3,4,5,6,7,8], where: b i1 is the mean absolute error of the i-th observer, b i2 is the maximum absolute error of the i-th observer, b i3is the average relative error of the i-th observer, b i4 is the coefficient of determination of the i-th observer, b i5 is the maximum convergence time of the i-th observer, b i6 is the average convergence time of the i-th observer, b i7 is the maximum chattering amplitude of the i-th observer, b i8 is the average chattering amplitude of the i-th observer;

[0021] The calculation of each evaluation index is as follows:

[0022] The mean absolute error of the i-th observer is:

[0023] The maximum absolute error of the i-th observer:

[0024] The average relative error of the i-th observer is:

[0025] The coefficient of determination of the i-th observer is:

[0026] The maximum convergence time of the i-th observer: b i5 =max{t i};

[0027] The average convergence time of the i-th observer is:

[0028] The maximum chattering amplitude of the i-th observer:

[0029] The average chattering amplitude of the i-th observer:

[0030] Where s is the number of samples, c is the location of data collection, and y is the actual intake pressure value. is the estimated value of the i-th observer, is the average value of the actual value; t i The time required for the estimated value of each data collection of the i-th observer to enter the error band of ±δ% of the actual value; is the maximum value of the estimated value of the i-th observer, is the minimum value of the estimated value of the i-th observer.

[0031] Furthermore, the numerical standardization of the evaluation indicators is as follows:

[0032]

[0033] Where: b ij is the jth evaluation index value of the i-th observer; b min and bmax They represent the minimum and maximum values ​​of the j-th evaluation index of all observers respectively; h ij is the standardized value of the jth evaluation indicator of the i-th observer;

[0034] Normalize the values ​​after the evaluation index is standardized:

[0035]

[0036] Among them, h ij is the standardized value of the jth evaluation index value of the i-th observer; l ij is the normalized value of the jth evaluation index of the i-th observer.

[0037] Furthermore, the entropy weight method is used to determine the weights of the evaluation indicators, as follows:

[0038] Calculate the entropy value e of the jth evaluation index in the comprehensive evaluation system j , the calculation formula is as follows:

[0039]

[0040] Where m is the total number of observers, l ij is the normalized value of the jth evaluation index of the i-th observer;

[0041] Calculate the weight w of the jth evaluation index according to the entropy value of the jth evaluation index j , the calculation formula is:

[0042]

[0043] Furthermore, the observer score is calculated by linear weighting. The specific steps are as follows:

[0044] Based on the weight of each evaluation value, the comprehensive score of the observer is obtained by linear weighted summation:

[0045]

[0046] Where: f i represents the comprehensive score of the i-th observer, l ij is the normalized value of the jth evaluation index of the i-th observer.

[0047] The beneficial effects of the present invention are:

[0048] 1. The hydrogen engine intake pressure estimation method described in the present invention divides the operating mode according to the throttle opening of the hydrogen engine, establishes a comprehensive evaluation index system for the observer under each operating condition, and determines the optimal observer selection strategy. This solves the problem of the inability to measure the intake pressure of the hydrogen engine due to the easy damage of the pressure sensor in the intake duct due to backfire, realizes accurate estimation of the intake pressure under different operating conditions, and ensures the normal operation of the hydrogen engine.

[0049] 2. The hydrogen engine intake pressure estimation method described in the present invention constructs a comprehensive observer evaluation matrix that includes multi-dimensional indicators such as mean absolute error and maximum absolute error, and standardizes and normalizes the evaluation indicators. Combined with the entropy weight method to calculate the indicator weights, it can comprehensively and objectively evaluate the observer performance, provide a scientific basis for the selection of the optimal observer, and improve the reliability and accuracy of the intake pressure estimation.

[0050] 3. The hydrogen engine intake pressure estimation method described in the present invention, which is based on the method of adaptively selecting the optimal observer under operating conditions, has strong versatility and scalability. It is not only applicable to various observers such as the extended Kalman filter observer, the unscented Kalman filter observer, and the sliding mode observer, but can also provide a reference for the estimation and control of other parameters in the field of hydrogen engine control technology, which will help promote the further development of hydrogen engine control technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. The drawings described below are some embodiments of the present invention. For ordinary technicians in this field, it is obvious that other drawings can be obtained based on these drawings without paying any creative work.

[0052] Picture 1 This is a schematic diagram of the working conditions of the hydrogen engine described in the present invention.

[0053] Picture 2 This is a flow chart of the method for estimating the intake pressure of a hydrogen engine according to the present invention. DETAILED DESCRIPTION

[0054] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and are not to be construed as limiting the present invention.

[0055] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "axial", "radial", "vertical", "horizontal", "inside", "outside" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0056] In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "connect," "fixed," etc. should be understood broadly. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediary; or internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0057] like Picture 2 As shown, the method for estimating the intake pressure of a hydrogen engine according to the present invention comprises the following steps:

[0058] S01: Working mode division, according to the throttle opening of the hydrogen engine, the working mode is divided into light load condition, medium load condition and heavy load condition, such as Picture 1 The specific classification standards are as follows:

[0059] When the throttle opening satisfies θ≤θ small.max When , the hydrogen engine is in a low-load condition;

[0060] When the throttle opening satisfies θ small.max <θ≤θ med.max When , the hydrogen engine is in medium load condition;

[0061] When the throttle opening satisfies θ med.max When <θ≤100%, the hydrogen engine is in a high-load condition;

[0062] Where: θ represents the throttle opening; θ small.max The maximum throttle opening allowed under low load conditions; θ med.max It is the maximum throttle opening allowed under medium load conditions.

[0063] S02: Establish a comprehensive evaluation index system, which includes mean absolute error (MAE), maximum absolute error (Max AE), mean relative error (MRE), coefficient of determination (R-squared), maximum convergence time (MCT), average convergence time (ACT), maximum jitter amplitude (MJA) and average jitter amplitude (AJA).

[0064] Among these evaluation metrics, MAE and Max AE directly quantify the absolute deviation between the estimated and actual values. MAE reflects the overall error level, while Max AE reflects the worst-case error scenario of the observer. MRE measures error as a relative ratio, making it suitable for horizontal comparisons across different pressure ranges, eliminating dimensionality effects and ensuring comparable error assessments. R-squared statistically evaluates the overall fit between the estimated and actual values. Values ​​closer to 1 indicate stronger model predictive capabilities. MCT and ACT serve as convergence speed evaluation metrics. MCT measures the maximum time it takes for the observer to reach a stable state, ensuring the responsiveness of real-time control. ACT reflects the average level of the observer's convergence speed, avoiding interference from single extreme values. MJA and AJA quantify the oscillation amplitude of the estimated value. MJA quantifies the maximum amplitude of the estimated value oscillation. Excessive chattering can lead to misjudgment by the control algorithm. AJA assesses the average strength of the estimated value oscillation, assisting in determining the long-term stability of the observer.

[0065] Data collection and processing: Run the hydrogen engine under different working conditions, collect the actual value of the intake pressure and the estimated values ​​of each observer in real time every T time, and calculate the value of each evaluation index.

[0066] The value of the j evaluation index of the i-th observer is denoted as b ij , j∈[1,2,3,4,5,6,7,8], where: b i1 is the mean absolute error of the i-th observer, b i2 is the maximum absolute error of the i-th observer, b i3 is the average relative error of the i-th observer, b i4 is the coefficient of determination of the i-th observer, b i5 is the maximum convergence time of the i-th observer, b i6 is the average convergence time of the i-th observer, b i7 is the maximum chattering amplitude of the i-th observer, b i8 is the average chattering amplitude of the i-th observer.

[0067] The calculation of each evaluation index is as follows:

[0068] Mean Absolute Error (MAE):

[0069]

[0070] Maximum absolute error (Max AE):

[0071]

[0072] Mean relative error (MRE):

[0073]

[0074] Coefficient of determination (Rs):

[0075]

[0076] Where s is the number of samples, c is the location of data collection, and y is the actual intake pressure value. is the estimated value of the i-th observer, is the average of the actual values;

[0077] Maximum Convergence Time (MCT):

[0078] b i5 =max{t i};

[0079] Average convergence time (ACT):

[0080]

[0081] Maximum jitter amplitude (MJA):

[0082]

[0083] Average jitter amplitude (AJA):

[0084]

[0085] Among them, t i The time required for the estimated value of each data collection of the i-th observer to enter the error band of ±δ% of the actual value;

[0086] is the maximum value of the estimated value of the i-th observer, is the minimum value of the estimated value of the i-th observer.

[0087] S03: Standardize and normalize the numerical values ​​of the evaluation indicators, as follows:

[0088]

[0089] Where: b ij is the jth evaluation index value of the i-th observer; b min and bmax They represent the minimum and maximum values ​​of the j-th evaluation index of all observers respectively; h ij is the standardized value of the jth evaluation indicator of the i-th observer;

[0090] Normalize the values ​​after the evaluation index is standardized:

[0091]

[0092] Among them, h ij The standardized value of the jth evaluation index value of the i-th observer; k ij is the normalized value of the jth evaluation index of the i-th observer.

[0093] S04: Determine the weights of the evaluation indicators using the entropy weight method, as follows:

[0094] Calculate the entropy value e of the jth evaluation index in the comprehensive evaluation system j , the calculation formula is as follows:

[0095]

[0096] Where m is the total number of observers, l ij is the normalized value of the jth evaluation index of the i-th observer;

[0097] Calculate the weight w of the jth evaluation index according to the entropy value of the jth evaluation index j , the calculation formula is:

[0098]

[0099] S05: Calculate the observer score through linear weighting. The specific steps are as follows:

[0100] Based on the weight of each evaluation value, the comprehensive score of the observer is obtained by linear weighted summation:

[0101]

[0102] Where: f i represents the comprehensive score of the i-th observer.

[0103] S06: Selection of observer;

[0104] According to the comprehensive scores of each observer, the one with the highest comprehensive score is selected as the optimal observer, and its estimated value is selected as the intake pressure output of the hydrogen engine.

[0105] Example

[0106] Taking a hydrogen engine as an example, the embodiment uses three observers, namely the extended Kalman filter observer (EKF), the unscented Kalman filter observer (UKF), and the sliding mode observer (SMO), as examples to describe the specific implementation methods under low load conditions and high load conditions. For medium load conditions, the implementation logic and technical paths of these two conditions can be referred to, and the optimal observer can be selected according to the operating characteristics of the corresponding conditions. The hydrogen engine intake pressure estimation method includes the following steps:

[0107] S01: Working mode classification: Based on the throttle opening of the hydrogen engine, the working mode is divided into light load condition, medium load condition and heavy load condition. The specific classification criteria are as follows:

[0108] When the throttle opening θ≤25%, the hydrogen engine is in a low-load condition;

[0109] When the throttle opening satisfies 25%<θ≤75%, the hydrogen engine is in medium load condition;

[0110] When the throttle opening satisfies 70%<θ≤100%, the hydrogen engine is in a high-load condition;

[0111] S02: Establish a comprehensive evaluation index system, operate the hydrogen engine under different working conditions, collect the actual value of the intake pressure and the estimated values ​​of each observer in real time every T time, and calculate the value of each evaluation index.

[0112] During a certain time period T under low-load conditions, s groups of actual intake pressure values ​​and estimated values ​​of each observer are collected. Based on the following calculation formulas for each evaluation index, the values ​​of each evaluation index of the extended Kalman filter observer, the unscented Kalman filter observer, and the sliding mode observer in this time period are calculated respectively.

[0113] The value of the j evaluation index of the i-th observer is denoted as b ij , j∈[1,2,3,4,5,6,7,8], i∈[1,2,3]; where: i=1 is the extended Kalman filter observer, i=2 is the unscented Kalman filter observer, and i=3 is the sliding mode observer; b i1 is the mean absolute error of the i-th observer, b i2 is the maximum absolute error of the i-th observer, b i3 is the average relative error of the i-th observer, b i4 is the coefficient of determination of the i-th observer, b i5 is the maximum convergence time of the i-th observer, b i6 is the average convergence time of the i-th observer, b i7 is the maximum chattering amplitude of the i-th observer, b i8 is the average jitter amplitude of the i-th observer. 11Expressed as the mean absolute error of the extended Kalman filter observer;

[0114] Mean Absolute Error (MAE):

[0115]

[0116] Maximum absolute error (Max AE):

[0117]

[0118] Mean relative error (MRE):

[0119]

[0120] Coefficient of determination (Rs):

[0121]

[0122] Where s is the number of samples, c is the location of data collection, and y is the actual intake pressure value. is the estimated value of the i-th observer, is the average of the actual values;

[0123] Maximum Convergence Time (MCT):

[0124] b i5 =max{t i};

[0125] Average convergence time (ACT):

[0126]

[0127] Maximum jitter amplitude (MJA):

[0128]

[0129] Average jitter amplitude (AJA):

[0130]

[0131] Among them, t i The time required for the estimated value of each data collection of the i-th observer to enter the error band of ±δ% of the actual value;

[0132] is the maximum value of the estimated value of the i-th observer, is the minimum value of the estimated value of the i-th observer.

[0133] The calculation results are shown in Table 1:

[0134] Table 1: Values ​​of various evaluation indicators under low load conditions

[0135]

[0136] S03: The numerical values ​​of the evaluation indicators are standardized and normalized, and the standardized results are shown in Table 2, which are as follows:

[0137]

[0138] Where: b ij is the jth evaluation index value of the i-th observer; b min and b max They represent the minimum and maximum values ​​of the j-th evaluation index of all observers respectively; h ij is the standardized value of the jth evaluation indicator of the i-th observer;

[0139] Table 2: Standardized values ​​of various evaluation indicators under low load conditions

[0140] <![CDATA[h i1 ]]> <![CDATA[h i2 ]]> <![CDATA[h i3 ]]> <![CDATA[h i4 ]]> <![CDATA[h i5 ]]> <![CDATA[h i6 ]]> <![CDATA[h i7 ]]> <![CDATA[h i8 ]]> EKF 0.5 0.286 0.286 0.625 1 1 1 1 UKF 0 0 0 1 0.833 0.75 0.571 0.667 SMO 1 1 1 0 0 0 0 0

[0141] Normalize the values ​​after the evaluation index is standardized:

[0142]

[0143] Among them, h ij is the standardized value of the jth evaluation index value of the i-th observer; l ij is the normalized value of the jth evaluation index of the i-th observer.

[0144] The normalized results are shown in Table 3:

[0145] Table 3: Normalized values ​​of evaluation indicators under low load conditions

[0146] <![CDATA[l i1 ]]> <![CDATA[l i2 ]]> <![CDATA[l i3 ]]> <![CDATA[l i4 ]]> <![CDATA[l i5 ]]> <![CDATA[l i6 ]]> <![CDATA[l i7 ]]> <![CDATA[l i8 ]]> EKF 0.333 0.222 0.222 0.385 0.546 0.571 0.637 0.6 UKF 0 0 0 0.615 0.454 0.429 0.363 0.4 SMO 0.667 0.778 0.778 0 0 0 0 0

[0147] S04: Determine the weights of the evaluation indicators using the entropy weight method, as follows:

[0148] Calculate the entropy value e of the jth evaluation index in the comprehensive evaluation system j , the calculation formula is as follows:

[0149]

[0150] Where m is the total number of observers, l ij is the normalized value of the jth evaluation index of the i-th observer;

[0151] Calculate the weight w of the jth evaluation index according to the entropy value of the jth evaluation index j , the calculation formula is:

[0152]

[0153] The weights of each evaluation index are shown in Table 4:

[0154] Table 4: Weights of various evaluation indicators under low load conditions

[0155] MAE Max AE MRE Rs MCT ACT MJA AJA <![CDATA[w j ]]> 0.126 0.155 0.155 0.117 0.098 0.113 0.121 0.116

[0156] S05: Calculate the observer score through linear weighting. The specific steps are as follows:

[0157] Based on the weight of each evaluation value, the comprehensive score of the observer is obtained by linear weighted summation:

[0158]

[0159] Where: f i represents the comprehensive score of the i-th observer.

[0160] The calculation results are shown in Table 5 below:

[0161] Table 5: Comprehensive scores of each observer under low load conditions

[0162] EKF UKF SMO <![CDATA[f i ]]> 0.420 0.255 0.325

[0163] S06: Selection of observer;

[0164] According to the comprehensive scores of each observer, the results in Table 5 show that the EKF has the highest comprehensive score, so its estimated value is used as the intake pressure output under low load conditions.

[0165] The specific implementation method under high load conditions is as follows:

[0166] S02: During a certain period T under high-load conditions, the actual intake pressure value and the estimated values ​​of each observer are collected in real time to obtain s groups of sample data. The obtained data is then analyzed and processed according to the numerical calculation formulas of each evaluation index, and the specific values ​​of each evaluation index are finally obtained, as shown in Table 6:

[0167] Table 6: Values ​​of various evaluation indicators under heavy load conditions

[0168]

[0169] S03: The numerical values ​​of the evaluation indicators are standardized and normalized, and the standardized results are shown in Table 7:

[0170] Table 7: Standardized values ​​of various evaluation indicators under heavy load conditions

[0171] <![CDATA[h i1 ]]> <![CDATA[h i2 ]]> <![CDATA[h i3 ]]> <![CDATA[h i4 ]]> <![CDATA[h i5 ]]> <![CDATA[h i6 ]]> <![CDATA[h i7 ]]> <![CDATA[h i8 ]]> EKF 0.333 0.333 0.2 0.6 1 1 0.364 0.4 UKF 0 0 0 1 0.333 0.5 0 0 SMO 1 1 1 0 0 0 1 1

[0172] The values ​​of the standardized evaluation indicators are normalized, and the normalized results are shown in Table 8:

[0173] Table 8: Normalized values ​​of evaluation indicators under heavy load conditions

[0174] <![CDATA[l i1 ]]> <![CDATA[l i2 ]]> <![CDATA[l i3 ]]> <![CDATA[l i4 ]]> <![CDATA[l i5 ]]> <![CDATA[l i6 ]]> <![CDATA[l i7 ]]> <![CDATA[l i8 ]]> EKF 0.25 0.25 0.167 0.375 0.75 0.667 0.267 0.286 UKF 0 0 0 0.625 0.25 0.333 0 0 SMO 0.75 0.75 0.833 0 0 0 0.733 0.714

[0175] S04: Determine the weight of the evaluation index by the entropy weight method, and obtain the weight of each evaluation index as shown in Table 9:

[0176] Table 9: Weights of various evaluation indicators under heavy load conditions

[0177] MAE Max AE MRE Rs MCT ACT MJA AJA <![CDATA[w j ]]> 0.128 0.128 0.155 0.105 0.128 0.111 0.124 0.120

[0178] S05: Calculate the observer score through linear weighting. The calculation results are shown in Table 10 below:

[0179] Table 10: Comprehensive scores of each observer under heavy load conditions

[0180] EKF UKF SMO <![CDATA[f i ]]> 0.367 0.135 0.498

[0181] S06: Selection of observer;

[0182] According to the comprehensive scores of each observer, the results obtained in Table 10 above show that SMO has the highest comprehensive score under high load conditions, so its estimated value is output as the intake pressure.

[0183] It should be understood that although this specification is described according to various embodiments, not every embodiment contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0184] The series of detailed descriptions listed above are only specific descriptions of feasible embodiments of the present invention. They are not intended to limit the scope of protection of the present invention. Any equivalent embodiments or changes that do not deviate from the technical spirit of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for estimating the intake pressure of a hydrogen engine, characterized in that: The steps include: Working mode classification: According to the throttle opening of the hydrogen engine, the working mode is divided into light load condition, medium load condition and heavy load condition; Run the hydrogen engine under different working conditions, collect the actual value of the intake pressure and the estimated value of each observer at every T time, and calculate the value of each evaluation index, which is recorded as b ij , b ij represents the value of the jth evaluation indicator of the i-th observer; the values ​​of the evaluation indicators include mean absolute error, maximum absolute error, mean relative error, determination coefficient, maximum convergence time, average convergence time, maximum jitter amplitude and average jitter amplitude; Standardization and normalization of the numerical values ​​of evaluation indicators; The weight of each evaluation indicator is determined by the entropy weight method; The comprehensive score of the observer is calculated by linear weighting; According to the comprehensive scores of each observer, the observer with the highest comprehensive score is selected as the optimal observer, and its estimated value is selected as the intake pressure output of the hydrogen engine.

2. The method for estimating the intake pressure of a hydrogen engine according to claim 1, wherein: The operating modes of hydrogen engines are divided into light load, medium load and heavy load conditions according to the throttle opening. The specific classification criteria are as follows: When the throttle opening satisfies θ≤θ small.max When , the hydrogen engine is in a low-load condition; When the throttle opening satisfies θ small.max <θ≤θ med.max When , the hydrogen engine is in medium load condition; When the throttle opening satisfies θ med.max When <θ≤100%, the hydrogen engine is in a high-load condition; Where: θ represents the throttle opening; θ small.max The maximum throttle opening allowed under low load conditions; θ med.max It is the maximum throttle opening allowed under medium load conditions.

3. The method for estimating the intake pressure of a hydrogen engine according to claim 1, wherein: The value of the i-th evaluation index of the i-th observer is denoted as b ij , j∈[1, 2, 3, 4, 5, 6, 7, 8], where: b i1 is the mean absolute error of the i-th observer, b i2 is the maximum absolute error of the i-th observer, b i3 is the average relative error of the i-th observer, b i4 is the coefficient of determination of the i-th observer, b i5 is the maximum convergence time of the i-th observer, b i6 is the average convergence time of the i-th observer, b i7 is the maximum chattering amplitude of the i-th observer, b i8 is the average chattering amplitude of the i-th observer; The calculation of each evaluation index is as follows: The mean absolute error of the i-th observer is: The maximum absolute error of the i-th observer: The average relative error of the i-th observer is: The coefficient of determination of the i-th observer is: The maximum convergence time of the i-th observer: b i5 =max{t i }; The average convergence time of the i-th observer is: The maximum chattering amplitude of the i-th observer: The average chattering amplitude of the i-th observer: Where s is the number of samples, c is the location of data collection, and y is the actual intake pressure value. is the estimated value of the i-th observer, is the average value of the actual value; t i The time required for the estimated value of each data collection of the i-th observer to enter the error band of ±δ% of the actual value; is the maximum value of the estimated value of the i-th observer, is the minimum value of the estimated value of the i-th observer.

4. The method for estimating the intake pressure of a hydrogen engine according to claim 1, wherein: The numerical standardization of evaluation indicators is as follows: Where: b ij is the jth evaluation index value of the i-th observer; b min and b max They represent the minimum and maximum values ​​of the j-th evaluation index of all observers respectively; h ij is the standardized value of the jth evaluation indicator of the i-th observer; Normalize the values ​​after the evaluation index is standardized: Among them, h ij is the standardized value of the jth evaluation index value of the i-th observer; l ij is the normalized value of the jth evaluation index of the i-th observer.

5. The method for estimating the intake pressure of a hydrogen engine according to claim 1, wherein: The weights of the evaluation indicators are determined by the entropy weight method, as follows: Calculate the entropy value e of the jth evaluation index in the comprehensive evaluation system j , the calculation formula is as follows: Where m is the total number of observers, l ij is the normalized value of the jth evaluation index of the i-th observer; Calculate the weight w of the jth evaluation index according to the entropy value of the jth evaluation index j , the calculation formula is:

6. The method for estimating the intake pressure of a hydrogen engine according to claim 1, characterized in that: The observer score is calculated by linear weighting. The specific steps are as follows: Based on the weight of each evaluation value, the comprehensive score of the observer is obtained by linear weighted summation: Where: f i represents the comprehensive score of the i-th observer, l ij is the normalized value of the jth evaluation index of the i-th observer.

Images