Diesel-Ignition Methanol Dual-Fuel Tractor Blending Ratio Control Method

By predicting future operating condition changes and combining Markov chains and genetic-ant colony algorithms to optimize the fuel supply control of diesel-ignited methanol dual-fuel tractors, the problems of response lag and inaccurate operating condition matching are solved, and timely adjustment of combustion state and efficient fuel utilization are achieved.

CN123109892APending Publication Date: 2026-11-17NORTHWEST A & F UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611595707.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-15
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

The existing fuel supply control method for diesel-ignited methanol dual-fuel tractors has a response lag problem, which makes it impossible to adjust the fuel supply ratio in advance before changes in operating conditions. Furthermore, it does not fully consider the dual effects of tractor speed and power output shaft torque, resulting in inaccurate combustion state matching.

Method used

By predicting future operating condition trends, a working condition model is constructed using Markov chains and a genetic-ant colony nested clustering algorithm. Combined with vehicle speed and power output shaft torque parameters, the injection parameters of diesel and methanol injectors, including injection pulse width, injection timing, and injection pressure, are adjusted in real time to achieve the fusion optimization of feedforward and feedback control quantities.

Benefits of technology

It enables timely and precise adjustment of fuel supply and injection parameters under varying operating conditions, improves the accuracy of combustion state matching and methanol substitution rate, reduces NOx and soot emissions, and shortens the response time after sudden changes in operating conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123109892A_ABST
    Figure CN123109892A_ABST
Patent Text Reader

Abstract

This invention provides a method for controlling the blending ratio of a diesel-ignited methanol dual-fuel tractor, belonging to the field of internal combustion engine fuel supply and injection control technology. The method involves real-time acquisition of tractor speed and power output shaft torque, constructing a dual-feature parameter system for both driving and operation, and performing dimensionality reduction. A genetic-ant colony nested clustering algorithm is used to obtain the operating condition model. Future operating condition sequences are predicted based on Markov chains. Based on this, a feedforward injection control quantity is determined, and a feedback injection control quantity is determined by combining the blending ratio deviation. Finally, the target blending ratio is obtained by fusing the combustion state injection correction quantity. The electronic control unit (ECU) controls the injection pulse width, injection timing, and injection pressure of the diesel and methanol injectors accordingly, and performs rolling optimization in each control cycle. This invention enables timely adjustment of fuel supply and injection parameters of a dual-fuel engine under varying operating conditions, effectively improving the methanol substitution rate and ensuring combustion stability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of internal combustion engine fuel supply and injection control technology, and in particular to a method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors. Background Technology

[0002] The fuel supply system of a diesel-ignited methanol dual-fuel engine typically includes a diesel injector, a methanol injector, and an electronic control unit (ECU). Its basic working principle is as follows: the methanol injector injects methanol into the intake manifold, allowing it to mix thoroughly with air to form a homogeneous premixed air-fuel mixture; the diesel injector directly injects a small amount of diesel fuel into the cylinder as ignition fuel to ignite the premixed methanol mixture. This combustion mode combines the characteristics of lean combustion and low-temperature combustion, significantly reducing soot and NOx emissions while maintaining power and improving fuel economy. The ECU controls the injection pulse width, injection timing, and injection pressure of the diesel and methanol injectors according to the target blending ratio, thereby adjusting the diesel-to-methanol supply ratio and achieving precise control of the dual-fuel engine's fuel supply. Applying this technology to tractors is of great significance for promoting the green transformation of agricultural machinery.

[0003] Currently, there is a certain research foundation on fuel supply control methods for diesel-methanol dual-fuel engines. For example, some methods determine the current engine load using throttle pedal position sensor signals, control the throttle actuator or diesel injector to adjust the diesel injection quantity, and thus control the methanol injection quantity; other methods acquire real-time operating condition data such as engine speed, load, and temperature, and combine them with fuel characteristic difference models to determine the optimal diesel to methanol supply ratio under the current operating conditions; still other methods use exhaust temperature and excess air coefficient to predict the maximum blending ratio, so as to achieve real-time adjustment of the blending ratio according to operating conditions. In addition, in the field of hybrid tractors, research has applied operating condition prediction methods to energy management strategies, such as energy management of hybrid tractors based on short-time vehicle speed prediction and model predictive control. All of the above methods involve fuel supply and injection control of dual-fuel engines, but have not yet organically combined operating condition prediction with the blending ratio and injection parameter control of diesel-ignited methanol dual-fuel tractors.

[0004] However, existing technologies still have the following shortcomings: First, most existing dual-fuel engine fuel supply control methods rely on feedback control based on current sensor signals (speed, load, throttle pedal position, etc.), which is a reactive adjustment method. In field scenarios where tractor operating conditions fluctuate frequently, adjusting the injection pulse width, injection timing, and injection pressure of diesel and methanol injectors only after the operating conditions have changed results in a response lag. This makes it impossible to adjust the fuel supply ratio in advance before the operating conditions change, making it difficult to achieve timely matching of optimal combustion conditions.

[0005] Second, existing control methods are mostly based on bench tests or road condition development, failing to fully consider the core characteristics that distinguish tractors from road vehicles, namely their operational characteristics, which are simultaneously affected by both vehicle speed and PTO (power take-off shaft) torque. When a tractor is operating in the field, its dynamic load changes are jointly determined by its vehicle speed and PTO torque. Relying solely on engine speed, load, or throttle pedal position is insufficient to accurately characterize its actual operating conditions, thus affecting the accuracy of fuel supply and injection control in dual-fuel engines.

[0006] Third, although operating condition prediction methods have been applied to the energy management of hybrid tractors, no one has yet introduced them into the fuel supply control and fuel injection control of diesel-ignited methanol dual-fuel tractors. The fuel supply system of dual-fuel tractors needs to adjust the injection quantity and parameters of diesel and methanol in real time according to field operating conditions. The injection pulse width, injection timing, and injection pressure of the diesel and methanol injectors directly affect the blending ratio and combustion state. Existing fuel supply control methods mostly rely on feedback adjustment based on sensor signals at the current moment, resulting in response lag and difficulty in adjusting injection parameters in advance before changes in operating conditions.

[0007] Therefore, there is an urgent need for a method that can combine the prediction of dual operating conditions of tractor driving and operation with the fuel supply control of dual-fuel engine. By predicting the trend of operating condition changes in the future control time domain in advance, the target blending ratio can be determined, and the electronic control unit can control the injection pulse width, injection timing and injection pressure of diesel injector and methanol injector accordingly, so as to realize the timely and accurate adjustment of fuel supply and injection parameters of diesel-ignited methanol dual-fuel tractor under changing operating conditions. Summary of the Invention

[0008] The purpose of this invention is to provide a method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors. By predicting future operating conditions, the target blending ratio is determined, and the electronic control unit controls the injection pulse width, injection timing, and injection pressure of the diesel and methanol injectors accordingly, thereby achieving timely adjustment of fuel supply and injection parameters of the dual-fuel engine.

[0009] To achieve the above objectives, the present invention provides a method for controlling the blending ratio of a diesel-ignited methanol dual-fuel tractor. The dual-fuel engine of the tractor includes a diesel injector, a methanol injector, and an electronic control unit, comprising the following steps: Step S1: Collect tractor operating status parameters, preprocess the collected data, divide the working condition segments and extract feature parameters to establish a working condition model of the tractor. Step S2: Establish a state transition probability matrix based on the working condition model, identify the current working condition category according to the current tractor operating state parameters, and use a prediction algorithm to predict the predicted working condition sequence in the future control time domain. Step S3: Based on the predicted operating condition sequence, the deviation between the current actual blending ratio and the target blending ratio, and the real-time combustion status of the engine, determine the target blending ratio. The target blending ratio is used to characterize the supply ratio of diesel and methanol. Step S31: Determine the feedforward injection control quantity based on the predicted operating condition sequence, determine the feedback injection control quantity based on the deviation between the current actual combustion ratio and the target combustion ratio, and determine the combustion state injection correction quantity based on the real-time combustion state of the engine. Step S32: Use the combustion state injection correction amount to correct the feedforward injection control amount and the feedback injection control amount, and fuse the corrected feedforward injection control amount, the corrected feedback injection control amount and the combustion state injection correction amount to obtain the target blending ratio. Step S33: The electronic control unit controls the injection parameters of the diesel injector and the methanol injector according to the target blending ratio to adjust the fuel supply of the dual-fuel engine. The injection parameters include injection pulse width, injection timing and injection pressure. Step S4: Repeat steps S2 and S3 in each control cycle, and use the injection parameters adjusted in step S33 as the initial value of the target blending ratio in step S31 of the next control cycle to realize the rolling optimization of dual-fuel injection parameters of the fuel supply system.

[0010] Preferably, in step S1, the tractor operating status parameters include tractor speed and power output shaft torque; the characteristic parameters include driving characteristic parameters and operation characteristic parameters. The driving characteristic parameters include average speed, maximum speed, speed standard deviation, acceleration time ratio, deceleration time ratio, and idling time ratio. The operation characteristic parameters include average power output shaft torque, maximum power output shaft torque, power output shaft torque standard deviation, torque fluctuation coefficient, and average power output shaft power. The working condition model contains multiple working condition categories and standard working condition characteristics for each category. The multiple working condition categories include road transport working condition, field transfer working condition, rotary tillage light load working condition, rotary tillage heavy load working condition, and plowing working condition. The rotary tillage light load working condition and rotary tillage heavy load working condition belong to the rotary tillage operation mode, and the plowing working condition belongs to the plowing operation mode. The two are determined and locked by the characteristic parameters or implement type signal when entering the operation. They are mutually exclusive operation modes and cannot be directly switched between each other.

[0011] Preferably, the clustering algorithm in step S1 is a genetic-ant colony nested clustering algorithm, specifically including: First-level clustering: The sample points in the principal component score matrix that contains both vehicle speed information and power output shaft torque information after dimensionality reduction are used as clustering objects. A genetic algorithm is used for global clustering with the optimization objective of minimizing intra-cluster distance and maximizing inter-cluster distance. After iteration until convergence, the preliminary clustering results and cluster centers are obtained. The second layer of clustering: The cluster centers output by the genetic algorithm are used as the initial pheromone distribution of the ant colony algorithm. The ant colony algorithm is used for local clustering. After iteration until convergence, the final clustering result is obtained. In the final clustering result, each cluster corresponds to a tractor driving-operation combination working condition category.

[0012] Preferably, step S2 specifically includes the following steps: Step S21: Define the various working condition categories obtained in step S1 as discrete state spaces. ,in For the first Similar working conditions, The total number of working condition categories is given; the frequency of transitions between each type of working condition is counted. Direct transitions between rotary tillage and plowing working conditions are not counted as valid transitions, or their state transition probabilities are set to zero and re-normalized row by row; the state transition probability matrix is ​​calculated. : Step S22: Real-time acquisition of the current moment speed and power output shaft torque The current working condition category is identified by matching it with the standard working conditions of each working condition category obtained in step S1. ,in For the current moment The operating condition category; at the same time, the current operating mode is determined and locked based on the characteristic parameters or implement type signal when entering the operation; when the current operating mode is rotary tillage mode, the subsequent status candidates only include rotary tillage operating conditions; when the current operating mode is plowing mode, the subsequent status candidates only include plowing operating conditions; the lock is only released when the field transfer or parking and implement change status is detected. Step S23: Based on the Markov chain, according to the current working conditions... and state transition probability matrix Predicting future control time domain The operating conditions at each time point are used to obtain the predicted operating condition sequence. ,in For predicting time intervals, , To predict the number of steps, For the first The working condition category at each prediction time; the working mode constraint is applied during prediction: when the current working condition is rotary tillage, the plowing working condition will not appear in the predicted working condition sequence; when the current working condition is plowing, the rotary tillage working condition will not appear in the predicted working condition sequence.

[0013] Preferably, control time domain The predicted time interval is 5 to 30 seconds. The time is 0.5 to 2 seconds; when the current working condition is identified in step S22 When the operation is rotary tillage or plowing, the control time domain is... Take 15-30 seconds; when the current working condition is identified in step S22 For road transport operating conditions, control time domain Take 5 to 15 seconds.

[0014] Preferably, the feedforward injection control quantity in step S31 is the weighted average of the optimal blending ratio at each predicted time point in the predicted operating condition sequence obtained in step S23, calculated according to the following formula: ; in, This is the feedforward injection control quantity. For the first The weight coefficients for each prediction time point, and , As the attenuation factor, For the first The optimal blending ratio corresponding to each predicted time point; When the current working condition is identified in step S22 When the operation is rotary tillage or plowing, Take a value of 0.1 to 0.3; when the current working condition is identified in step S22 When operating under road transport conditions, Take a value of 0.3 to 0.5.

[0015] Preferably, in step S31, the feedback injection control quantity is calculated using a proportional-integral-derivative control algorithm according to the following formula: ; in, To provide feedback on injection control quantity, This represents the deviation between the actual blending ratio and the target blending ratio at the current moment. This is the proportionality coefficient. The integral coefficient is... These are the differential coefficients; When the calcination ratio deviates When the content is greater than 20% of the target blending ratio, It is 1.0~1.5. The value is between 0.01 and 0.05. The value is between 0.001 and 0.01. When the calcination ratio deviates When it is less than 5% of the target blending ratio, The value is 0.3~0.6. The value is 0.08~0.15. The value is 0.01~0.05; When the calcination ratio deviates When the target blending ratio is between 5% and 20%, The value is 0.6~1.0. The value is 0.05~0.08. It ranges from 0.005 to 0.02.

[0016] Preferably, the combustion state injection correction amount in step S31 is determined as follows: Real-time combustion state parameters of the engine are collected, including knock intensity, combustion cycle variation coefficient, and exhaust temperature. The combustion cycle variation coefficient is denoted as... ; Based on knock intensity and combustion cycle variation coefficient Calculate the knock health index based on exhaust temperature and other parameters. Combustion stability index thermal state index Calculate the combustion state correction factor according to the following formula. As a correction amount for injection under combustion conditions: ; in, This is a correction factor for combustion state. This is the weighting coefficient for the knockout health index. The weighting coefficients for the combustion stability index are... This is the weighting coefficient for the thermal state index.

[0017] Preferably, in step S32, a combustion state correction coefficient is used. The feedforward injection control quantity and the feedback injection control quantity are corrected according to the following formulas: ; ; in, This is the corrected feedforward injection control value. This is the corrected feedback injection control value. This is the feedforward injection control quantity. To provide feedback on injection control quantity, For reference correction factor, This is the correction intensity coefficient for the feedforward injection control quantity. The correction intensity coefficient for the feedback injection control quantity; when When the correction is positive, the feedforward injection control quantity and the feedback injection control quantity are amplified; when When the time comes, the correction is reversed to suppress both the feedforward injection control and the feedback injection control. In step S32, the corrected feedforward injection control quantity, the corrected feedback injection control quantity, and the combustion state correction coefficient are combined. The target blending ratio is obtained by fusing according to the following formula: ; in, For the target blending ratio, The fusion weighting coefficients for the corrected feedforward injection control quantity. The fusion weighting coefficients for the corrected feedback injection control quantity. The fusion weighting coefficient is the combustion state correction coefficient. When the current working condition is identified in step S22 When performing rotary tillage operations, It is 0.35-0.45. It is 0.25-0.35. It is 0.25-0.35; When the current working condition is identified in step S22 When the plowing operation is in progress, It is 0.35-0.45. It is 0.25-0.35. It is 0.25-0.35; When the current working condition is identified in step S22 When operating under road transport conditions, It is 0.45-0.55. It is 0.20-0.30. It is 0.20-0.30.

[0018] Preferably, in step S33, the target blending ratio obtained in step S32 is used. Amplitude limiting is applied; The limiting process determines the safety boundary as follows: when the engine knock intensity exceeds the knock threshold, the target combustion ratio is adjusted. Set to the minimum safe blending ratio corresponding to the current operating conditions; When the engine exhaust temperature exceeds the temperature threshold, the target blending ratio will be adjusted. Set to the minimum safe blending ratio corresponding to the current operating conditions; When the engine combustion cycle variation coefficient When the stability threshold is exceeded, the target doping ratio will be adjusted. Set to the minimum safe blending ratio corresponding to the current operating conditions; Based on the target blending ratio after the limit is applied, the diesel injection quantity and methanol injection quantity are calculated using the following formula: ; ; in, This refers to the diesel injection quantity. This refers to the methanol injection volume. This is the equivalent amount of total fuel energy. Because of the low calorific value of diesel fuel, It has the low calorific value of methanol; Based on the calculated diesel injection quantity and methanol injection volume The electronic control unit adjusts the injection parameters of the diesel and methanol injectors, including injection pulse width, injection timing, and injection pressure, to control the fuel supply of the dual-fuel engine.

[0019] Therefore, the present invention employs the above-mentioned method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors, and the beneficial technical effects are as follows: (1) This invention introduces a working condition prediction method based on Markov chains to know the trend of working condition changes in the future control time domain in advance, and uses the feedforward injection control quantity as a component of the target blending ratio, so that the electronic control unit can adjust the injection parameters of the diesel injector and methanol injector in advance before the working condition changes. This solves the problem of response lag caused by the existing dual-fuel engine fuel supply control method relying only on the sensor signal at the current moment for feedback control, and realizes timely adjustment of fuel supply and injection parameters of dual-fuel engine under tractor working condition changes.

[0020] (2) This invention constructs a dual characteristic parameter system that combines driving characteristic parameters and operation characteristic parameters by simultaneously collecting vehicle speed and power output shaft torque as input parameters for working condition modeling. This enables the constructed working condition model to reflect the operating characteristics of the tractor under the dual influence of driving speed and power output shaft torque. It provides accurate working condition basis for determining the target blending ratio and controlling the injection parameters of diesel injectors and methanol injectors. This solves the problem that existing fuel supply control methods cannot accurately characterize the dynamic load changes of tractors in the field because they do not fully consider the dual operating characteristics of tractors.

[0021] (3) This invention introduces a combustion state correction coefficient obtained by weighting three sub-indicators: knock intensity, combustion cycle variation coefficient, and exhaust temperature. This correction coefficient is used to perform bidirectional correction on the feedforward injection control quantity and the feedback injection control quantity. Safety limits are set according to three trigger conditions: knock intensity, exhaust temperature, and combustion cycle variation coefficient. At the same time, differentiated values ​​of control time domain, feedforward attenuation factor, and fusion weight coefficient are set for different operating conditions. This improves the methanol substitution rate while ensuring the combustion stability of the engine. Meanwhile, by converting the target blending ratio into the injection pulse width, injection timing, and injection pressure of the diesel and methanol injectors, the electronic control unit drives the diesel and methanol injectors to adjust the dual fuel supply, thereby realizing the precise control of the dual fuel injection parameters by the fuel supply system. Attached Figure Description

[0022] Figure 1 This is a flowchart of the method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors according to the present invention; Figure 2 To continuously optimize the target blending ratio variation curve; Figure 3 For comparison of the co-firing ratio response curves (at the moment of sudden change in operating conditions) =30s). Detailed Implementation

[0023] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0024] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0025] Example 1 This embodiment uses a 180-horsepower diesel-ignited methanol dual-fuel engine mounted on a tractor to implement the fuel supply control method of this invention under typical paddy field rotary tillage conditions. Figure 1 As shown, the method includes: Step S1: Collect tractor operating status parameters. After preprocessing the collected data, divide it into operating condition segments and extract feature parameters. After dimensionality reduction of the feature parameters, perform cluster analysis using a clustering algorithm to obtain the tractor's operating condition model, providing a basis for calculating the target blending ratio for fuel supply control.

[0026] A sensor system is installed on the tractor, including: a GNSS speed measurement system for acquiring vehicle speed signals at a sampling frequency of 10Hz; a dynamic torque sensor mounted on the power take-off shaft for acquiring power take-off shaft torque signals at a sampling frequency of 10Hz; and a multi-channel data acquisition system for synchronously acquiring and storing signals from each sensor. The engine's fuel supply system includes diesel injectors and methanol injectors. The diesel injectors are direct injection injectors, and the methanol injectors are port injectors. Both are connected to the electronic control unit (ECU), which controls their injection pulse width, injection timing, and injection pressure according to the target blending ratio. Field operation data is continuously collected for 8 hours.

[0027] The collected raw data were preprocessed as follows: median filtering was used to remove outliers, low-pass filtering was used for noise reduction, and linear interpolation was used to imput missing values.

[0028] The preprocessed continuous data was divided into operating condition segments according to the complete process of "stop → start → run → stop", resulting in 1568 valid operating condition segments. Feature parameters for each operating condition segment were extracted, including 6 driving feature parameters (average vehicle speed, maximum vehicle speed, vehicle speed standard deviation, acceleration time ratio, deceleration time ratio, and idling time ratio) and 5 operational feature parameters (average torque of the power output shaft, maximum torque of the power output shaft, standard deviation of the power output shaft torque, torque fluctuation coefficient, and average power of the power output shaft), as well as the duration of the operating condition segment, forming a 12-dimensional feature parameter matrix.

[0029] Principal component analysis (PCA) is performed on the feature parameter matrix: first, variable standardization is performed, then the correlation coefficient matrix is ​​calculated, followed by solving for the eigenvalues ​​and eigenvectors, and finally, the contribution rate and cumulative contribution rate of each principal component are calculated. The top four principal components with a cumulative contribution rate of 85% are selected as the dimensionality-reduced principal component score matrix.

[0030] A nested genetic-ant colony clustering algorithm was used to perform cluster analysis on the principal component score matrix. The first layer used the sample points in the principal component score matrix as the clustering objects, employing a genetic algorithm for global clustering. The genetic algorithm had a population size of 100, a crossover probability of 0.8, a mutation probability of 0.05, and a maximum of 200 iterations. The optimization objective was to minimize intra-cluster distance and maximize inter-cluster distance. After convergence, preliminary clustering results and cluster centers were obtained. The second layer used the cluster centers output by the genetic algorithm as the initial pheromone distribution for the ant colony algorithm, employing a local clustering algorithm. The ant colony algorithm had 50 ants, a pheromone evaporation coefficient of 0.3, and a maximum of 100 iterations. After convergence, the final clustering results were obtained.

[0031] Cluster analysis revealed that the 1568 work condition segments were divided into 5 clusters. The work condition categories and mean characteristic parameters of each cluster are shown in Table 1. Among them, rotary tillage with light load and rotary tillage with heavy load belong to the rotary tillage operation mode; plowing belongs to the plowing operation mode. Both are determined and locked by the characteristic parameters or implement type signals when entering the operation.

[0032] Table 1. Mean values ​​of characteristic parameters and optimal blending ratios for various operating conditions

[0033] Step S2: Establish a state transition probability matrix based on the working condition model, identify the current working condition category according to the current tractor operating state parameters, and use a prediction algorithm to predict the predicted working condition sequence in the future control time domain.

[0034] Step S21: Define the various working condition categories obtained in step S1 as discrete state spaces. ,in For the first Similar working conditions, The total number of working condition categories is given; the frequency of transitions between each type of working condition is counted. Direct transitions between rotary tillage and plowing working conditions are not counted as valid transitions, or their state transition probabilities are set to zero and re-normalized row by row; the state transition probability matrix is ​​calculated. : ; in, To be based on working conditions Transfer to working condition The probability, To be based on working conditions Transfer to working condition Number of times, For working conditions Total number of occurrences.

[0035] In this embodiment, the five types of working conditions are defined as discrete state spaces. ={Road transport, light rotary tillage, heavy rotary tillage, plowing, field transfer}. The frequency of transfers between various working conditions was statistically analyzed, and the state transition probability matrix was calculated. The results are shown in Table 2.

[0036] Table 2 State transition probability matrix

[0037] Step S22: Real-time acquisition of the current moment speed and power output shaft torque The current working condition category is identified by matching it with the standard working conditions of each working condition category obtained in step S1. ,in For the current moment The operating condition category. At the same time, the current operating mode is determined and locked based on the characteristic parameters or implement type signals when entering the operation; in rotary tillage mode, plowing conditions are not predicted, and in plowing mode, rotary tillage conditions are not predicted; the lock is only released when field transfer or parking and implement changing status is detected.

[0038] In this embodiment, the vehicle speed =5.8km / h and power output shaft torque =495 N·m, matched with the standard operating conditions for each operating condition category in Table 1, to identify the current operating condition category. This is for heavy-duty rotary tillage conditions.

[0039] Step S23: Based on the Markov chain, according to the current working conditions... and state transition probability matrix Predicting future control time domain The operating conditions at each time point are used to obtain the predicted operating condition sequence. ,in For predicting time intervals, , To predict the number of steps, For the first The operating condition category at each predicted time point, When making predictions, the following work mode constraints are applied: if the current work condition is rotary tillage, the predicted work condition sequence will not include plowing; if the current work condition is plowing, the predicted work condition sequence will not include rotary tillage.

[0040] Control Time Domain The predicted time interval is 5 to 30 seconds. The time is 0.5 to 2 seconds; when the current working condition is identified in step S22 When the operation is rotary tillage or plowing, the control time domain is... Take 15-30 seconds; when the current working condition is identified in step S22 For road transport operating conditions, control time domain Take 5 to 15 seconds.

[0041] In this embodiment, the control time domain is set. =20 seconds (current working condition is rotary tillage), predicted time interval =1 second, number of prediction steps =20. Based on Markov chains, according to the current operating conditions. = Rotary tillage heavy load and the state transition probability matrix in Table 2 Predict the operating conditions at each time point within the next 20 seconds to obtain the predicted operating condition sequence. The prediction results are as follows: seconds 1-8 are heavy rotary tillage, seconds 9-14 are light rotary tillage, and seconds 15-20 are heavy rotary tillage.

[0042] Step S3: Adaptively control the blending ratio based on the predicted operating condition sequence, the deviation between the current actual blending ratio and the target blending ratio, and the real-time combustion status of the engine. Step S31: Determine the feedforward injection control quantity based on the predicted operating condition sequence, determine the feedback injection control quantity based on the deviation between the current actual combustion ratio and the target combustion ratio, and determine the combustion state injection correction quantity based on the real-time combustion state of the engine.

[0043] Based on the predicted operating condition sequence obtained in step S23, the optimal blending ratio corresponding to each operating condition is looked up in Table 1 to obtain the optimal blending ratio sequence for each predicted time (unit: %): to ={38,38,38,38,38,38,38,38,26,26,26,26,26,26,38,38,38,38,38}.

[0044] Calculate the feedforward injection control quantity using the following formula: ; in, This is the feedforward injection control quantity. For the first The optimal blending ratio corresponding to each predicted time point For the first The weight coefficients for each prediction time point, and , The attenuation factor is 0 < <1.

[0045] It should be noted that, The following rules apply: When the current operating condition is identified in step S22... When the operation is rotary tillage or plowing, Take a value of 0.1 to 0.3; when the current working condition is identified in step S22 When operating under road transport conditions, Take a value of 0.3 to 0.5.

[0046] In this embodiment, the current operating condition This is a heavy-duty rotary tillage operation, therefore... Take 0.2.

[0047] Weighting coefficients for each prediction time point The sequence is: {0.819,0.670,0.549,0.449,0.368,0.301,0.247,0.202,0.165,0.135,0.111,0.091,0.074,0.061,0.050,0.041,0.033,0.027,0.022,0.018}.

[0048] Calculated feedforward injection control quantity =31.8%.

[0049] Real-time monitoring shows the current actual blending ratio is 28.5%, while the current target blending ratio is 32.0%, indicating a deviation. =-3.5% (meaning the actual blending ratio is 3.5 percentage points lower than the target blending ratio), blending ratio deviation =3.5% / 32.0%=10.9%.

[0050] It should be noted that the proportional coefficient in the feedback injection control quantity... Integral coefficient Differential coefficients Configure according to the following rules: When the calcination ratio deviates When the content is greater than 20% of the target blending ratio, It is 1.0~1.5. The value is between 0.01 and 0.05. The value is between 0.001 and 0.01. When the calcination ratio deviates When it is less than 5% of the target blending ratio, The value is 0.3~0.6. The value is 0.08~0.15. The value is 0.01~0.05; When the calcination ratio deviates When the target blending ratio is between 5% and 20%, The value is 0.6~1.0. The value is 0.05~0.08. It ranges from 0.005 to 0.02.

[0051] In this embodiment, =10.9%, which falls between 5% and 20% of the target blending ratio, therefore Take 0.8, Take 0.06, Take 0.01.

[0052] The proportional-integral-derivative (PI-DE) control algorithm is used to calculate the feedback injection control quantity according to the following formula: ; in, To provide feedback on injection control quantity, This represents the deviation between the actual blending ratio and the target blending ratio at the current moment. At the current moment, the cumulative value of the integral term is 1.2%, and the differential term is -0.3%. Therefore: =-2.7%.

[0053] Real-time combustion status parameters of the engine were collected: the current knock intensity is 0.12 MPa (knock threshold is 0.25 MPa), which is 60% below the knock threshold, and the knock health index is [not specified]. Set to 1.0; Current combustion cycle variation coefficient The combustion stability index is 3.5%. The value is set to 0.85; the current exhaust temperature is 385℃, which is within the optimal operating temperature range (350~420℃), and the thermal state index is... Take 1.0.

[0054] Set weight coefficients =0.4, =0.3, =0.3, calculate the combustion state correction factor according to the following formula. As a correction amount for injection under combustion conditions: ; in, This is a correction factor for combustion state. This is the weighting coefficient for the knockout health index. The weighting coefficients for the combustion stability index are... is the weighting coefficient of the thermal state index, and .

[0055] Step S32: Use the combustion state injection correction amount to correct the feedforward injection control amount and the feedback injection control amount, and fuse the corrected feedforward injection control amount, the corrected feedback injection control amount and the combustion state injection correction amount to obtain the target blending ratio.

[0056] Using combustion state correction factor The feedforward injection control quantity and the feedback injection control quantity are corrected according to the following formulas: ; ; in, This is the corrected feedforward injection control value. This is the corrected feedback injection control value. This is the feedforward injection control quantity. To provide feedback on injection control quantity, For reference correction factor, The value is typically 0.7-0.8; in this embodiment, it is set to 0.75. This is the correction intensity coefficient for the feedforward injection control quantity. The value should be between 0.2 and 0.5; in this embodiment, it is set to 0.35. The correction intensity coefficient for the feedback injection control quantity. The value is typically between 0.3 and 0.6; in this embodiment, it is set to 0.45. > ; when When the correction is positive, the feedforward injection control quantity and the feedback injection control quantity are amplified; when When the time comes, the correction is reversed to suppress both the feedforward injection control and the feedback injection control.

[0057] because =0.955> =0.75, corrected to a positive correction.

[0058] The corrected feedforward injection control, the corrected feedback injection control, and the combustion state correction coefficient are combined. The target blending ratio is obtained by fusing according to the following formula: ; in, For the target blending ratio, The fusion weighting coefficients for the corrected feedforward injection control quantity. The fusion weighting coefficients for the corrected feedback injection control quantity. This is the fusion weighting coefficient for the combustion state correction coefficient, and .

[0059] When the current working condition is identified in step S22 When performing rotary tillage operations, It is 0.35-0.45. It is 0.25-0.35. It is 0.25-0.35; When the current working condition is identified in step S22 When the plowing operation is in progress, It is 0.35-0.45. It is 0.25-0.35. It is 0.25-0.35; When the current working condition is identified in step S22 When operating under road transport conditions, It is 0.45-0.55. It is 0.20-0.30. It is 0.20-0.30.

[0060] In this embodiment, the current operating condition This is a heavy-duty rotary tillage operation, therefore... =0.40, =0.30, =0.30.

[0061] Calculate the target blending ratio: =41.42%.

[0062] Step S33: The electronic control unit adjusts the diesel injection parameters and methanol injection parameters according to the target blending ratio.

[0063] Based on the target calcination ratio obtained in step S32 Amplitude limiting is applied.

[0064] The limiting process determines the safety boundary as follows: when the engine knock intensity exceeds the knock threshold, the target combustion ratio is adjusted. Set as the minimum safe blending ratio corresponding to the current operating conditions.

[0065] When the engine exhaust temperature exceeds the temperature threshold, the target blending ratio will be adjusted. Set as the minimum safe blending ratio corresponding to the current operating conditions.

[0066] When the engine combustion cycle variation coefficient When the stability threshold is exceeded, the target doping ratio will be adjusted. Set as the minimum safe blending ratio corresponding to the current operating conditions.

[0067] Based on the target blending ratio after the limit is applied, the diesel injection quantity and methanol injection quantity are calculated using the following formula: ; ; in, This refers to the diesel injection quantity. This refers to the methanol injection volume. This is the equivalent amount of total fuel energy. Because of the low calorific value of diesel fuel, It has the low calorific value of methanol.

[0068] Based on the calculated diesel injection quantity and methanol injection volume The electronic control unit adjusts the injection parameters of the diesel and methanol injectors, including injection pulse width, injection timing, and injection pressure.

[0069] In this embodiment, the current target doping ratio =41.42%, which is within the safe boundary (the safe blending ratio range for heavy-duty rotary tillage under current operating conditions is 30%~48%), and no limit is required.

[0070] Set the total fuel energy equivalent =100mg / cycle (based on diesel energy equivalent), diesel low calorific value =42.7 MJ / kg, methanol has a low calorific value =19.9MJ / kg.

[0071] =58.58mg / cycle; =88.89mg / cycle; Based on the calculated diesel and methanol injection quantities, adjust the diesel injection pulse width to 580 μs, the injection timing to 6°A before top dead center, adjust the methanol injection pulse width to 1240 μs, the injection timing to 60°A after the intake valve opens, and the injection pressure to 4.5 bar.

[0072] Step S4: Repeat steps S2 and S3 in each control cycle, and use the injection parameters adjusted in step S33 as the initial value of the target blending ratio in step S31 of the next control cycle.

[0073] Steps S2 and S3 are repeated in each control cycle (1 second). The initial target blending ratio for the next control cycle is the actual blending ratio of 41.42% achieved after adjustment in the current cycle. Continuous rolling optimization is performed as the operating conditions change, and the target blending ratio change curves for each control cycle are shown below. Figure 2 As shown.

[0074] Under the same rotary tillage conditions, a comparative experiment was conducted between this method and the traditional PID feedback control method based on the current working conditions. The results are shown in Table 3.

[0075] Table 3 Performance comparison between this method and the PID feedback control method

[0076] exist Figure 3 middle, =30 is the point of abrupt change in operating conditions (from light rotary tillage to heavy rotary tillage). This method is approximately at... =35 (i.e., 5 seconds after the mutation) to reach a stable doping ratio of 40%, while the traditional PID method takes approximately... =38.5 seconds (i.e. 8.5 seconds after the mutation) to reach a stable doping ratio of 36%, the response time of this method is shortened by about 40%.

[0077] In this embodiment, under rotary tillage conditions, the diesel-ignited methanol dual-fuel tractor, after adopting this method, achieved the following: an average methanol substitution rate of 36.6%, which is 8.3 percentage points higher than the traditional PID feedback control method (28.3%); a 6.7% reduction in equivalent fuel consumption; a 12.4% reduction in NOx emissions; an 18.6% reduction in soot emissions; and a reduction in the blending ratio response time after sudden changes in operating conditions from 8.5 seconds to 5.1 seconds, a reduction of approximately 40%. The engine did not exhibit knocking or misfire during 8 hours of continuous operation, demonstrating good combustion stability.

[0078] It is worth noting that all contents not described in detail in this invention are existing technologies and are well known to those skilled in the art.

[0079] Therefore, the present invention adopts the above-mentioned diesel-ignited methanol dual-fuel tractor blending ratio control method, which can effectively solve the problems of delayed response of blending ratio control and insufficient consideration of the dual operation characteristics of tractors in the prior art. It is applicable to the dynamic adaptive control of the blending ratio of dual-fuel engines under tractor field operation conditions.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for controlling the blending ratio of a diesel-ignited methanol dual-fuel tractor, wherein the dual-fuel engine of the tractor includes a diesel injector, a methanol injector, and an electronic control unit, characterized in that... Includes the following steps: Step S1: Collect tractor operating status parameters, preprocess the collected data, divide the working condition segments and extract feature parameters to establish a working condition model of the tractor. Step S2: Establish a state transition probability matrix based on the working condition model, identify the current working condition category according to the current tractor operating state parameters, and use a prediction algorithm to predict the predicted working condition sequence in the future control time domain. Step S3: Based on the predicted operating condition sequence, the deviation between the current actual blending ratio and the target blending ratio, and the real-time combustion status of the engine, determine the target blending ratio. The target blending ratio is used to characterize the supply ratio of diesel and methanol. Step S31: Determine the feedforward injection control quantity based on the predicted operating condition sequence, determine the feedback injection control quantity based on the deviation between the current actual combustion ratio and the target combustion ratio, and determine the combustion state injection correction quantity based on the real-time combustion state of the engine. Step S32: Use the combustion state injection correction amount to correct the feedforward injection control amount and the feedback injection control amount, and fuse the corrected feedforward injection control amount, the corrected feedback injection control amount and the combustion state injection correction amount to obtain the target blending ratio. Step S33: The electronic control unit controls the injection parameters of the diesel injector and the methanol injector according to the target blending ratio to adjust the fuel supply of the dual-fuel engine. The injection parameters include injection pulse width, injection timing and injection pressure. Step S4: Repeat steps S2 and S3 in each control cycle, and use the injection parameters adjusted in step S33 as the initial value of the target blending ratio in step S31 of the next control cycle to realize the rolling optimization of dual-fuel injection parameters of the fuel supply system.

2. The method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors according to claim 1, characterized in that, In step S1, the tractor operating status parameters include tractor speed and power output shaft torque; the characteristic parameters include driving characteristic parameters and operation characteristic parameters. The driving characteristic parameters include average speed, maximum speed, speed standard deviation, acceleration time ratio, deceleration time ratio, and idling time ratio. The operation characteristic parameters include average power output shaft torque, maximum power output shaft torque, power output shaft torque standard deviation, torque fluctuation coefficient, and average power output shaft power. The working condition model contains multiple working condition categories and standard working condition characteristics for each category. The multiple working condition categories include road transport working condition, field transfer working condition, rotary tillage light load working condition, rotary tillage heavy load working condition, and plowing working condition. The rotary tillage light load working condition and rotary tillage heavy load working condition belong to the rotary tillage operation mode, and the plowing working condition belongs to the plowing operation mode. The two are determined and locked by the characteristic parameters or implement type signal when entering the operation. They are mutually exclusive operation modes and cannot be directly switched between each other.

3. The method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors according to claim 1, characterized in that, The clustering algorithm in step S1 is a genetic-ant colony nested clustering algorithm, which specifically includes: First-level clustering: The sample points in the principal component score matrix that contains both vehicle speed information and power output shaft torque information after dimensionality reduction are used as clustering objects. A genetic algorithm is used for global clustering with the optimization objective of minimizing intra-cluster distance and maximizing inter-cluster distance. After iteration until convergence, the preliminary clustering results and cluster centers are obtained. The second layer of clustering: The cluster centers output by the genetic algorithm are used as the initial pheromone distribution of the ant colony algorithm. The ant colony algorithm is used for local clustering. After iteration until convergence, the final clustering result is obtained. In the final clustering result, each cluster corresponds to a tractor driving-operation combination working condition category.

4. The method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors according to claim 1, characterized in that, Step S2 specifically includes the following steps: Step S21: Define the various working condition categories obtained in step S1 as discrete state spaces. ,in For the first Similar working conditions, The total number of working condition categories is given; the frequency of transitions between each type of working condition is counted. Direct transitions between rotary tillage and plowing working conditions are not counted as valid transitions, or their state transition probabilities are set to zero and re-normalized row by row; the state transition probability matrix is ​​calculated. : Step S22: Real-time acquisition of the current moment speed and power output shaft torque The current working condition category is identified by matching it with the standard working conditions of each working condition category obtained in step S1. ,in For the current moment The operating condition category; at the same time, the current operating mode is determined and locked based on the characteristic parameters or implement type signal when entering the operation; when the current operating mode is rotary tillage mode, the subsequent status candidates only include rotary tillage operating conditions; when the current operating mode is plowing mode, the subsequent status candidates only include plowing operating conditions; the lock is only released when the field transfer or parking and implement change status is detected. Step S23: Based on the Markov chain, according to the current working conditions... and state transition probability matrix Predicting future control time domain The operating conditions at each time point are used to obtain the predicted operating condition sequence. ,in For predicting time intervals, , To predict the number of steps, For the first The working condition category at each prediction time; the working mode constraint is applied during prediction: when the current working condition is rotary tillage, the plowing working condition will not appear in the predicted working condition sequence; when the current working condition is plowing, the rotary tillage working condition will not appear in the predicted working condition sequence.

5. The method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors according to claim 4, characterized in that, Control Time Domain The predicted time interval is 5 to 30 seconds. The time is 0.5 to 2 seconds; when the current working condition is identified in step S22 When the operation is rotary tillage or plowing, the control time domain is... Take 15-30 seconds; when the current working condition is identified in step S22 For road transport operating conditions, control time domain Take 5 to 15 seconds.

6. The method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors according to claim 4, characterized in that, In step S31, the feedforward injection control quantity is the weighted average of the optimal blending ratio at each predicted time point in the predicted operating condition sequence obtained in step S23, calculated according to the following formula: ; in, This is the feedforward injection control quantity. For the first The weight coefficients for each prediction time point, and , As the attenuation factor, For the first The optimal blending ratio corresponding to each predicted time point; When the current working condition is identified in step S22 When the operation is rotary tillage or plowing, Take a value of 0.1 to 0.3; when the current working condition is identified in step S22 When operating under road transport conditions, Take a value of 0.3 to 0.

5.

7. The method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors according to claim 4, characterized in that, In step S31, the feedback injection control quantity is calculated using a proportional-integral-derivative control algorithm according to the following formula: ; in, To provide feedback on injection control quantity, This represents the deviation between the actual blending ratio and the target blending ratio at the current moment. This is the proportionality coefficient. The integral coefficient is... These are the differential coefficients; When the calcination ratio deviates When the content is greater than 20% of the target blending ratio, It is 1.0~1.

5. The value is between 0.01 and 0.

05. The value is between 0.001 and 0.

01. When the calcination ratio deviates When it is less than 5% of the target blending ratio, The value is 0.3~0.

6. The value is 0.08~0.

15. The value is 0.01~0.05; When the calcination ratio deviates When the target blending ratio is between 5% and 20%, The value is 0.6~1.

0. The value is 0.05~0.

08. It ranges from 0.005 to 0.

02.

8. The method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors according to claim 4, characterized in that, The combustion state injection correction amount in step S31 is determined as follows: Real-time combustion state parameters of the engine are collected, including knock intensity, combustion cycle variation coefficient, and exhaust temperature. The combustion cycle variation coefficient is denoted as... ; Based on knock intensity and combustion cycle variation coefficient Calculate the knock health index based on exhaust temperature and other parameters. Combustion stability index thermal state index Calculate the combustion state correction factor according to the following formula. As a correction amount for injection under combustion conditions: ; in, This is a correction factor for combustion state. This is the weighting coefficient for the knockout health index. The weighting coefficients for the combustion stability index are... This is the weighting coefficient for the thermal state index.

9. The method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors according to claim 1, characterized in that, In step S32, the combustion state correction factor is used. The feedforward injection control quantity and the feedback injection control quantity are corrected according to the following formulas: ; ; in, This is the corrected feedforward injection control value. This is the corrected feedback injection control value. This is the feedforward injection control quantity. To provide feedback on injection control quantity, For reference correction factor, This is the correction intensity coefficient for the feedforward injection control quantity. The correction intensity coefficient for the feedback injection control quantity; when When the correction is positive, the feedforward injection control quantity and the feedback injection control quantity are amplified; when When the time comes, the correction is reversed to suppress both the feedforward injection control and the feedback injection control. In step S32, the corrected feedforward injection control quantity, the corrected feedback injection control quantity, and the combustion state correction coefficient are combined. The target blending ratio is obtained by fusing according to the following formula: ; in, For the target blending ratio, The fusion weighting coefficients for the corrected feedforward injection control quantity. The fusion weighting coefficients for the corrected feedback injection control quantity. The fusion weighting coefficient is the combustion state correction coefficient. When the current working condition is identified in step S22 When performing rotary tillage operations, It is 0.35-0.

45. It is 0.25-0.

35. It is 0.25-0.35; When the current working condition is identified in step S22 When the plowing operation is in progress, It is 0.35-0.

45. It is 0.25-0.

35. It is 0.25-0.35; When the current working condition is identified in step S22 When operating under road transport conditions, It is 0.45-0.

55. It is 0.20-0.

30. It is 0.20-0.

30.

10. The method for controlling the blending ratio of diesel-ignited methanol dual-fuel tractors according to claim 1, characterized in that, In step S33, the target blending ratio obtained in step S32 is used. Amplitude limiting is applied; The limiting process determines the safety boundary as follows: when the engine knock intensity exceeds the knock threshold, the target combustion ratio is adjusted. Set to the minimum safe blending ratio corresponding to the current operating conditions; When the engine exhaust temperature exceeds the temperature threshold, the target blending ratio will be adjusted. Set to the minimum safe blending ratio corresponding to the current operating conditions; When the engine combustion cycle variation coefficient When the stability threshold is exceeded, the target doping ratio will be adjusted. Set to the minimum safe blending ratio corresponding to the current operating conditions; Based on the target blending ratio after the limit is applied, the diesel injection quantity and methanol injection quantity are calculated using the following formula: ; ; in, This refers to the diesel injection quantity. This refers to the methanol injection volume. This is the equivalent amount of total fuel energy. Because of the low calorific value of diesel fuel, It has the low calorific value of methanol; Based on the calculated diesel injection quantity and methanol injection volume The electronic control unit adjusts the injection parameters of the diesel and methanol injectors, including injection pulse width, injection timing, and injection pressure, to control the fuel supply of the dual-fuel engine.