Engine air intake amount control mechanism and engine
By introducing a variable intake manifold adjustment component and a machine learning model into the engine intake volume control mechanism, intelligent automatic adjustment of engine intake volume is achieved, solving the problems of throttle response speed and adjustment accuracy, and improving engine power output and fuel economy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2026-03-31
AI Technical Summary
Existing engine intake volume control mechanisms suffer from limited throttle response speed, especially during rapid acceleration or deceleration, making it difficult to quickly adjust to the optimal intake volume, which affects the smoothness and responsiveness of power output; the throttle body is prone to wear, increasing maintenance costs and failure rates; and traditional intake volume control strategies have limited accuracy in adjusting airflow, making it difficult to improve combustion efficiency and emission control levels.
The control unit, which combines a variable intake manifold adjustment component and a machine learning model, predicts the intake flow rate by collecting engine model, speed and throttle status parameters in real time and performs real-time compensation and adjustment through a flow sensor, thereby realizing intelligent automatic adjustment of the intake manifold gas flow rate.
It maintains high power output and fuel economy under various operating conditions, improves the engine's fuel economy at low and medium speeds and high power output at high speeds, and enhances intake efficiency and combustion efficiency.
Smart Images

Figure CN119084197B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of engine technology, and in particular to an engine intake air volume control mechanism and an engine. Background Technology
[0002] In modern automotive engine technology, precise control of intake air volume is crucial for improving engine performance, fuel economy, and reducing emissions. Currently, the most widely used engine intake air volume control mechanisms on the market are achieved through a throttle body in conjunction with an electronic control unit (ECU). This mechanism changes the amount of air entering the cylinder by adjusting the throttle opening.
[0003] However, this traditional approach has several significant drawbacks: First, the throttle response speed is limited, especially under rapid acceleration or deceleration conditions, making it difficult to quickly adjust to the optimal intake volume, affecting the smoothness and responsiveness of the engine's power output; second, the throttle is prone to wear during frequent opening and closing, increasing maintenance costs and failure rates; third, the precision of traditional intake volume control strategies in adjusting airflow is limited by the design of the mechanical structure, making it difficult to further improve the engine's combustion efficiency and emission control levels. Summary of the Invention
[0004] The purpose of this invention is to overcome the limitations of existing throttle response speeds, particularly under rapid acceleration or deceleration conditions, making it difficult to quickly adjust to the optimal intake volume, thus affecting the smoothness and responsiveness of engine power output. Secondly, the throttle body is prone to wear during frequent opening and closing, increasing maintenance costs and failure rates. Furthermore, the accuracy of traditional intake volume control strategies in adjusting airflow is limited by the design of the mechanical structure, making it difficult to further improve engine combustion efficiency and emission control levels. This invention provides an engine intake volume control mechanism and engine, with a specially developed control unit for the engine. Through the inclusion of a model training module, a predictive intake volume control algorithm is trained. The machine learning model for intake manifold airflow collects engine model, engine speed, and throttle status parameters in real time. Based on the trained machine learning model, it predicts the intake manifold airflow. Then, the intake airflow in the intake manifold is collected in real time by a flow sensor and labeled as L1. It is compared and analyzed with L to generate a compensation flow L2, where L2 = L - L1. When L2 > 0, an intake compensation command is generated and sent to the control terminal. When L2 ≤ 0, no intake compensation command is generated. This realizes intelligent automatic adjustment of intake manifold gas flow, enabling the engine to maintain high power output and fuel economy under various operating conditions.
[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:
[0006] This invention discloses an engine intake air volume control mechanism, comprising:
[0007] Air intake chamber;
[0008] An intake pipe is connected to the outer wall of the intake chamber and is used to supply air into the intake chamber.
[0009] An intake manifold is connected to the outer wall of the intake chamber and is used to deliver air from the intake chamber to the engine. Several intake manifolds are provided in total.
[0010] A plurality of variable intake manifold adjustment components are provided, and the plurality of variable intake manifold adjustment components are respectively disposed in the plurality of intake manifolds, for automatically adjusting the intake manifold flow cross section, thereby regulating the airflow velocity entering the engine through the intake manifold.
[0011] Preferably, the variable intake manifold adjustment assembly includes: a plurality of valve plates and valve stems, wherein the plurality of valve plates are fixed to the inner wall of the intake manifold in a ring-shaped distribution and the included angle between adjacent valve plates is equal, and a plurality of valve disc grooves are provided at equal intervals on the side wall of the valve plate away from the intake manifold.
[0012] The valve stem is rotatably located at the center of the intake manifold. Several valve discs are welded at equal intervals on the outer wall of the valve stem, and each valve disc corresponds to a valve disc groove. When the valve stem is driven to rotate, the valve discs can be rotated along the valve stem to adjust the overlapping area of the valve discs and the valve disc grooves.
[0013] Preferably, a bearing is fitted on the end of the valve stem away from the air intake chamber, a connecting rod is welded to the outer wall of the bearing, and the other end of the connecting rod is fixedly connected to the valve plate;
[0014] The valve stem has a linkage rod integrally formed at one end near the intake chamber, and the linkage rod extends into the intake chamber at the other end away from the valve stem and is welded with a linkage gear.
[0015] Preferably, the system also includes a drive mechanism for driving the valve stem to rotate and adjusting the overlapping area of the valve disc and the valve disc groove.
[0016] The drive mechanism includes a drive rod and a rack. The drive rod is slidably disposed on the inner wall of the air intake chamber. A plurality of racks are evenly spaced on the upper surface of the drive rod, and the racks are meshed with a linkage gear.
[0017] Preferably, the driving mechanism further includes a driving disk, and a connecting rod is rotatably connected to the corner of the upper surface of the driving disk, the other end of which is rotatably connected to the driving rod.
[0018] Preferably, a limiting sleeve is fitted on the drive rod between two adjacent racks. The limiting sleeve has a rectangular hollow structure and is fixed to the inner wall of the air intake chamber by welding with a connecting column.
[0019] Preferably, a connecting seat is provided on the outer wall of the intake chamber, and the intake manifold is fixed to the intake chamber by being threaded into the connecting seat. An installation hole is provided on the side wall of the intake manifold, and a flow sensor is provided on the outer wall of the intake manifold at the position of the installation hole.
[0020] Preferably, the system also includes a control unit, which comprises:
[0021] The data acquisition module is used to collect historical engine intake flow control data. The historical engine intake flow control data is collected when the engine fuel efficiency meets the target. The historical engine intake flow control data includes engine model, engine speed, throttle status and intake manifold intake flow.
[0022] The throttle valve status includes an open state and a closed state;
[0023] The model training module trains a machine learning model to predict the intake manifold airflow based on historical engine intake flow control data. It collects real-time engine model, engine speed, and throttle state parameters, and predicts the intake manifold airflow based on the trained machine learning model, which is labeled as L.
[0024] The intake compensation module collects the intake air flow in the intake manifold in real time through a flow sensor, which is marked as L1. It compares and analyzes L to generate a compensation flow L2, where L2 = L - L1. When L2 > 0, an intake compensation command is generated and sent to the control terminal. When L2 ≤ 0, no intake compensation command is generated.
[0025] Preferably, the specific method for training the machine learning model to predict the intake manifold airflow includes:
[0026] The collected historical engine intake airflow control data is converted into a corresponding set of feature vectors;
[0027] The engine model, engine speed, and throttle state parameters from historical engine intake air flow control data are used as inputs to a machine learning model. The machine learning model outputs the intake air flow corresponding to each set of engine model, engine speed, and throttle state parameters, and uses the actual intake air flow corresponding to each set of engine model, engine speed, and throttle state parameters as the prediction target. The training objective is to minimize the loss function value of the machine learning model. Training stops when the loss function value of the machine learning model is less than or equal to the preset target loss value.
[0028] An engine, including the aforementioned engine intake volume control mechanism.
[0029] Compared with the prior art, the present invention has the following beneficial effects:
[0030] During use, the cross-sectional area of the air passage in the intake manifold 3 can be dynamically and automatically adjusted based on the engine's intake air flow requirements. That is, at low speeds (i.e., when the intake air flow is low), by reducing the cross-sectional area of the intake passage, the gas flow velocity can be increased, thereby producing a better dynamic charging effect and improving intake efficiency (higher ηv). This adjustment helps to enhance the engine's intake speed at low and medium speeds, thereby increasing the airflow intensity in the cylinder, improving the combustion process, and improving the engine's fuel economy at low and medium speeds. At high speeds (i.e., when the intake air flow is high), increasing the cross-sectional area of the intake passage can reduce the gas flow resistance, which can also improve intake efficiency (higher ηv). This adjustment helps the engine achieve high power output at high speeds.
[0031] A control unit specifically developed for the engine is equipped with a model training module to train a machine learning model that predicts the intake manifold airflow. It collects engine model, engine speed, and throttle position parameters in real time. Based on the trained machine learning model, it predicts the intake manifold airflow. Then, a flow sensor collects the intake manifold airflow in real time, labeling it L1. This L1 is compared and analyzed with L to generate a compensation flow L2, where L2 = L - L1. When L2 > 0, an intake compensation command is generated and sent to the control terminal; when L2 ≤ 0, no intake compensation command is generated. This achieves intelligent automatic adjustment of the intake manifold gas flow, enabling the engine to maintain high power output and fuel economy under various operating conditions. Attached Figure Description
[0032] Figure 1 This is a schematic diagram of the overall structure of the present invention;
[0033] Figure 2 This is a side view of the present invention;
[0034] Figure 3 This is a cross-sectional view of the air intake chamber of the present invention;
[0035] Figure 4 This is a schematic diagram of the intake manifold in this invention;
[0036] Figure 5 This is a cross-sectional view of the intake manifold in this invention;
[0037] Figure 6 This is an exploded view of the variable intake manifold adjustment assembly in this invention;
[0038] Figure 7In this invention Figure 3 Enlarged view of area A in the image;
[0039] Figure 8 In this invention Figure 3 Enlarged view of area B in the image;
[0040] Figure 9 In this invention Figure 6 Enlarged view of area C in the image.
[0041] Reference numerals: 1. Intake chamber; 2. Intake pipe; 3. Intake manifold; 31. Variable intake duct adjustment assembly; 311. Valve disc; 312. Valve stem; 313. Linkage rod; 314. Linkage gear; 315. Valve flap; 316. Valve flap groove; 317. Bearing; 318. Connecting rod; 32. Drive mechanism; 321. Drive rod; 322. Rack; 323. Drive disc; 324. Linkage rod; 325. Limit sleeve; 326. Connecting column; 33. Mounting hole; 4. Connecting seat; 5. Flow sensor. Detailed Implementation
[0042] The following is in conjunction with the appendix Figure 1 - Appendix Figure 9 This invention further illustrates a specific implementation of an engine intake air volume control mechanism, overcoming the limitations of existing throttle response speeds, particularly during rapid acceleration or deceleration, which makes it difficult to quickly adjust to the optimal intake air volume, affecting the smoothness and responsiveness of engine power output. Secondly, the throttle valve is prone to wear during frequent opening and closing, increasing maintenance costs and failure rates. Furthermore, the traditional intake air volume control strategy's adjustment accuracy for airflow is limited by the mechanical structure design, making it difficult to further improve engine combustion efficiency and emission control levels. This invention provides an engine intake air volume control mechanism with a dedicated control unit developed for the engine, featuring a model training... The training module trains a machine learning model to predict the intake manifold airflow. It collects engine model, engine speed, and throttle status parameters in real time. Based on the trained machine learning model, it predicts the intake manifold airflow. Then, it collects the intake manifold airflow in real time through a flow sensor and marks it as L1. It compares and analyzes L1 with L to generate a compensation flow L2, where L2 = L - L1. When L2 > 0, an intake compensation command is generated and sent to the control terminal. When L2 ≤ 0, no intake compensation command is generated. This realizes intelligent automatic adjustment of the intake manifold gas flow, enabling the engine to maintain high power output and fuel economy under various operating conditions.
[0043] Example 1
[0044] Reference Figures 1-9 This embodiment provides an engine intake air volume control mechanism, including:
[0045] An engine intake air volume control mechanism, comprising:
[0046] Intake chamber 1;
[0047] The intake pipe 2 is connected to the outer wall of the intake chamber 1 and is used to supply air into the intake chamber 1;
[0048] The intake manifold 3 is connected to the outer wall of the intake chamber 1 and is used to deliver air from the intake chamber 1 to the engine. There are several intake manifolds 3 in total.
[0049] A number of variable intake manifold adjustment components 31 are provided. The variable intake manifold adjustment components 31 are respectively installed in a number of intake manifolds 3. They are used to automatically adjust the flow cross section of the intake manifolds 3 to regulate the airflow speed entering the engine through the intake manifolds 3.
[0050] The variable intake manifold adjustment assembly 31 includes: a plurality of valve plates 311 and valve stems 312. The plurality of valve plates 311 are fixed on the inner wall of the intake manifold 3 and are arranged in a ring. The included angle between two adjacent valve plates 311 is equal. A plurality of valve disc grooves 316 are provided at equal intervals on the side wall of the valve plate 311 away from the intake manifold 3.
[0051] The valve stem 312 is rotatably located at the center of the intake manifold 3. Several valve discs 315 are welded at equal intervals on the outer wall of the valve stem 312, and the valve discs 315 correspond one-to-one with the valve disc grooves 316. When the valve stem 312 is driven to rotate, the valve discs 315 can be driven to rotate along the valve stem 312, adjusting the overlapping area of the valve discs 315 and the valve disc grooves 316.
[0052] A bearing 317 is sleeved on the end of the valve stem 312 away from the intake chamber 1. A connecting rod 318 is welded on the outer wall of the bearing 317. The other end of the connecting rod 318 is fixedly connected to the valve plate 311.
[0053] The valve stem 312 has an integrally formed linkage rod 313 at the end near the intake chamber 1, and the end of the linkage rod 313 away from the valve stem 312 extends into the intake chamber 1 and is welded with a linkage gear 314.
[0054] Specifically, the variable intake manifold adjustment assembly 31 operates as follows: the drive mechanism drives the linkage gear 314 to rotate, the linkage gear 314 rotates, which in turn drives the valve stem 312 to rotate. The rotation of the valve stem 312 drives the valve disc 315 on the valve stem 312 to rotate. The rotation of the valve disc 315 changes the overlapping area between the valve disc 315 and the valve disc groove 316. It can be seen that the larger the overlapping area between the valve disc 315 and the valve disc groove 316, the larger the intake manifold flow cross section in the intake manifold 3, and vice versa.
[0055] In this embodiment, during use, the cross-sectional area of the air passage in the intake manifold 3 can be dynamically and automatically adjusted based on the engine's intake airflow requirements. That is, at low speeds (i.e., lower intake airflow), by reducing the cross-sectional area of the intake passage, the gas flow velocity can be increased, thereby producing a better dynamic charging effect and improving intake efficiency (higher ηv). This adjustment helps to enhance the engine's intake speed at low and medium speeds, thereby increasing the airflow intensity in the cylinder, improving the combustion process, and improving the engine's fuel economy at low and medium speeds. At high speeds (i.e., higher intake airflow), increasing the cross-sectional area of the intake passage can reduce the gas flow resistance, which can also improve intake efficiency (also higher ηv). This adjustment helps the engine achieve high power output at high speeds.
[0056] Example 2
[0057] Based on the above embodiments, this implementation further discloses the specific structure of the drive mechanism 32. In use, the drive mechanism 32 drives the variable intake manifold adjustment component 31 to work, thereby realizing dynamic and automatic adjustment of the cross-sectional area of the air flow channel in the intake manifold 3.
[0058] It also includes a drive mechanism 32, which is used to drive the valve stem 312 to rotate and adjust the overlapping area of the valve disc 315 and the valve disc groove 316.
[0059] The drive mechanism 32 includes a drive rod 321 and a rack 322. The drive rod 321 is slidably disposed on the inner wall of the air intake chamber 1. A plurality of racks 322 are evenly spaced on the upper surface of the drive rod 321. The racks 322 are meshed with the linkage gear 314.
[0060] The drive mechanism 32 also includes a drive disk 323, and a connecting rod 324 is rotatably connected to the corner of the upper surface of the drive disk 323. The other end of the connecting rod 324 is rotatably connected to the drive rod 321.
[0061] A limiting sleeve 325 is fitted on the drive rod 321 between two adjacent racks 322. The limiting sleeve 325 is a rectangular hollow structure. The limiting sleeve 325 is welded and fixed to the inner wall of the air intake chamber 1 by a connecting column 326.
[0062] In this embodiment, during use, the drive motor installed on the outer wall of the air intake chamber 1 drives the drive disk 323 to rotate. It should be noted that the drive motor is not shown in the figure. The rotation of the drive disk 323 pulls the connecting rod 324 to move. The movement of the connecting rod 324 pulls the drive rod 321 to move. The movement of the drive rod 321 drives the rack 322 to move. The movement of the rack 322 drives the linkage gear 314 to rotate.
[0063] Example 3
[0064] Based on the above embodiments, this embodiment further adds a control unit, which is used to automatically adjust the intake flow rate of the intake manifold 3;
[0065] A connecting seat 4 is provided on the outer wall of the intake chamber 1. The intake manifold 3 is fixed to the intake chamber 1 by being threaded into the connecting seat 4. An installation hole 33 is provided on the side wall of the intake manifold 3. A flow sensor 5 is provided on the outer wall of the intake manifold 3 at the position of the installation hole 33.
[0066] It also includes a control unit, which includes:
[0067] The data acquisition module is used to collect historical engine intake flow control data. The historical engine intake flow control data is collected under the condition that the engine fuel efficiency meets the target. The historical engine intake flow control data includes engine model, engine speed, throttle status and intake manifold 3 intake flow.
[0068] Throttle valve status includes open and closed states;
[0069] The model training module trains a machine learning model to predict the intake flow rate of intake manifold 3 based on historical engine intake flow control data. It collects real-time engine model, engine speed, and throttle state parameters, and predicts the intake flow rate of intake manifold 3 based on the trained machine learning model, which is labeled as L.
[0070] The intake compensation module uses flow sensor 5 to collect the intake airflow in the intake manifold 3 in real time, labeling it as L1. It compares and analyzes this L1 with L to generate a compensation flow rate L2, where L2 = L - L1. When L2 > 0, an intake compensation command is generated and sent to the control terminal; when L2 ≤ 0, no intake compensation command is generated. It should be noted that when L2 ≤ 0, it indicates sufficient intake airflow, therefore no intake compensation is needed; conversely, when L2 > 0, it indicates insufficient intake airflow, requiring intake compensation.
[0071] The specific methods for training a machine learning model to predict the intake airflow of intake manifold 3 include:
[0072] The collected historical engine intake airflow control data is converted into a corresponding set of feature vectors;
[0073] The engine model, engine speed, and throttle state parameters from historical engine intake air flow control data are used as inputs to the machine learning model. The machine learning model outputs the intake air flow corresponding to each set of engine model, engine speed, and throttle state parameters, and uses the actual intake air flow corresponding to each set of engine model, engine speed, and throttle state parameters as the prediction target. The training objective is to minimize the loss function value of the machine learning model. Training stops when the loss function value of the machine learning model is less than or equal to the preset target loss value.
[0074] It should be noted that the machine learning model can be either a deep neural network model or a deep belief network model, and the loss function of the machine learning model is the mean squared error.
[0075] Mean squared error is one of the commonly used loss functions. It is obtained by... The model is trained with minimization as the objective, which enables the first machine learning model to better fit the data, thereby improving the model's performance and accuracy.
[0076] In the loss function, MSE1 is the loss function value of the machine learning model, x is the feature vector group number; m is the number of feature vector groups; y x Let x be the intake airflow corresponding to the x-th eigenvector group. The intake flow rate corresponding to the x-th feature vector in real time;
[0077] Other model parameters of the machine learning model, such as the target loss value, optimization algorithm, ratio of training set to test set to validation set, and optimization of the loss function, are all obtained through actual engineering implementation and continuous experimental tuning.
[0078] In this embodiment, a control unit specifically developed for the engine is equipped with a model training module to train a machine learning model that predicts the intake airflow of the intake manifold 3. The engine model, engine speed, and throttle state parameters are collected in real time. Based on the trained machine learning model, the intake airflow of the intake manifold 3 is predicted. Then, the intake airflow in the intake manifold 3 is collected in real time by the flow sensor 5 and marked as L1. It is compared and analyzed with L to generate a compensation flow L2, where L2 = L - L1. When L2 > 0, an intake compensation command is generated and sent to the control terminal. When L2 ≤ 0, no intake compensation command is generated. This realizes the intelligent automatic adjustment of the intake manifold gas flow, enabling the engine to maintain high power output and fuel economy under various operating conditions.
[0079] Example 4
[0080] An engine, including the aforementioned engine intake volume control mechanism.
[0081] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.
Claims
1. An engine air intake amount control mechanism characterized by comprising: Include: Intake chamber (1); Intake pipe (2) is connected to the outer wall of the intake chamber (1), for conveying air into the intake chamber (1); Intake manifold (3), the intake manifold (3) is connected to the outer wall of the intake chamber (1), for conveying air in the intake chamber (1) to the engine, the intake manifold (3) is provided with a plurality of; Variable intake port adjustment assembly (31) is provided with a plurality of, a plurality of variable intake port adjustment assembly (31) is arranged in a plurality of intake manifold (3), for automatically adjusting the intake port flow cross section in the intake manifold (3), realizing the regulation of air flow rate through the intake manifold (3) into the engine; The variable intake port adjustment assembly (31) comprises: a plurality of valve pieces (311) and valve rods (312), a plurality of valve pieces (311) are fixedly connected to the inner wall of the intake manifold (3) in ring shape, and the included angle between adjacent two valve pieces (311) is equal, a plurality of valve pieces (311) are provided with a plurality of valve pieces (316) on the side wall away from the intake manifold (3) at equal intervals; The valve rod (312) is rotatably arranged at the inner center of the intake manifold (3), a plurality of valve petals (315) are welded on the outer wall of the valve rod (312) at equal intervals, and a plurality of valve petals (315) correspond to a plurality of valve petals (316) one by one, when the valve rod (312) is driven to rotate, the valve petals (315) can be driven to rotate along the valve rod (312), and the overlapping area of the valve petals (315) and the valve petals (316) is adjusted; Further comprising a driving mechanism (32), the driving mechanism (32) is used for driving the valve rod (312) to rotate, adjusting the overlapping area of the valve petals (315) and the valve petals (316); The driving mechanism (32) comprises a driving rod (321) and a rack (322), the driving rod (321) is slidably arranged on the inner wall of the intake chamber (1), a plurality of racks (322) are arranged on the upper surface of the driving rod (321) at equal intervals, and the rack (322) is connected with the linkage gear (314); The driving mechanism (32) further comprises a driving disc (323), the driving disc (323) is further rotatably connected with a linkage rod (324) at the corner of the upper surface, and the other end of the linkage rod (324) is rotatably connected with the driving rod (321); The outer wall of the intake chamber (1) is provided with a connecting seat (4), the intake manifold (3) is connected in the connecting seat (4) to realize the fixation between the intake manifold (3) and the intake chamber (1), the side wall of the intake manifold (3) is provided with a mounting hole (33), and the outer wall of the intake manifold (3) is provided with a flow sensor (5) at the position of the mounting hole (33); Further comprising a control unit, the control unit comprises: A data collection module is configured to collect historical engine intake flow control data, which is collected under the condition that engine fuel efficiency meets the standard, and the historical engine intake flow control data includes engine model, engine speed, throttle state, and intake manifold (3) intake flow; The throttle state includes an open state and a closed state; A model training module is configured to train a machine learning model for predicting the intake manifold (3) intake flow based on the historical engine intake flow control data, collect real-time engine model, engine speed, and throttle state parameters, and predict the intake manifold (3) intake flow based on the trained machine learning model, which is marked as L. The intake compensation module collects the intake flow in the intake manifold (3) in real time through the flow sensor (5), marked as L1, compares and analyzes it with L, and generates a compensation flow L2, wherein L2=L-L 1, When L2>0, an intake compensation instruction is generated and sent to the control terminal. When L2≤0, no intake compensation instruction is generated.
2. An engine air intake amount control mechanism according to claim 1, characterized by: A bearing (317) is sleeved on the end of the valve rod (312) away from the air chamber (1), a connecting rod (318) is welded on the outer wall of the bearing (317), and the other end of the connecting rod (318) is fixedly connected with the valve plate (311). The valve rod (312) is integrally formed with a linkage rod (313) at one end close to the air chamber (1), the linkage rod (313) extends into the air chamber (1) at an end away from the valve rod (312) and is welded with a linkage gear (314).
3. An engine air intake control mechanism according to claim 1, wherein: A limiting sleeve (325) is sleeved on the driving rod (321) between two adjacent racks (322), the limiting sleeve (325) is a rectangular hollow structure, and the limiting sleeve (325) and the inner wall of the air chamber (1) are fixedly connected by a connecting column (326).
4. An engine air intake control mechanism according to claim 1, characterized by: The specific method for training the machine learning model for predicting the intake manifold (3) intake flow includes: The collected historical engine intake flow control data is converted into a corresponding group of feature vectors; In the historical engine intake flow control data, the engine model, engine speed, and throttle state parameters are used as inputs of the machine learning model, the machine learning model takes the intake flow corresponding to each group of engine model, engine speed, and throttle state parameters as output, takes the actual intake flow corresponding to each group of engine model, engine speed, and throttle state parameters as a prediction target, and takes minimizing the machine learning model loss function value as a training target, and the training is stopped when the machine learning model loss function value is less than or equal to the preset target loss value.
5. An engine characterized by, An engine intake flow control mechanism is provided.
Citation Information
Patent Citations
Transient air flow control
CN103256132A
High-speed gasoline engine instantaneous condition air-fuel ratio control method
CN109184932A
Atomizer air inflow pipeline adjusting assembly for atomized metal powder production
CN110756817A
Variable vortex air intake manifold structure of engine
CN205605339U