An integrated hydraulic retarder valve module regulation control system

Abstract

This application discloses an integrated hydraulic retarder valve module adjustment and control system, belonging to the field of transmission control. It addresses the problems existing in the prior art. In this system, a pressure sensor detects the intake pressure of the hydraulic retarder. The intake end of the intake valve is connected in series with the air source, and the exhaust end is connected to the hydraulic retarder. A pressure relief valve, an exhaust valve, and a quick-release valve form a parallel exhaust valve group. The intake end of the parallel exhaust valve group is connected to the hydraulic retarder, and the exhaust end is connected to the atmosphere through the hydraulic retarder. The regulator is connected to the intake valve, pressure sensor, quick-release valve, and exhaust valve via a circuit. The control system is built using MATLAB / Simulink and calculates and decides on the control target through its own control strategy. The adjustment and control system of this application features guaranteed driving safety, sensitive and precise control, low manufacturing cost, simplicity and reliability, and long service life.

Description

Technical Field

[0001] This invention relates to an integrated hydraulic retarder valve module adjustment and control system, belonging to the field of image processing. Background Technology

[0002] A hydraulic retarder is an auxiliary braking device. In situations requiring prolonged or intense braking, such as long downhill slopes, traditional braking methods can lead to reduced braking efficiency or even brake failure due to brake overheating. Using a hydraulic retarder for auxiliary braking can effectively reduce brake usage and prevent brake fade. When the vehicle requires auxiliary braking, the retarder activates, using the counter-torque applied to the rotor by the retarder fluid to provide auxiliary braking for the vehicle.

[0003] When the retarder is working, retarder oil needs to be filled into the working chamber. However, when the retarder is not working, the retarder oil in the working chamber will generate resistance that hinders the movement of the car. Therefore, the retarder oil needs to be discharged from the working chamber when the retarder is not working. Usually, an intake and exhaust regulation system is used to use air pressure to force the retarder oil into and out of the retarder working chamber.

[0004] The magnitude of the braking torque generated by the retarder is related to the oil pressure and quantity in the working chamber and the rotor speed. The active adjustment of the retarder's braking torque mainly depends on the adjustment of the oil quantity in the retarder. However, due to the complex and varied driving conditions of heavy commercial vehicles, the retarder needs to work frequently and the power changes frequently. The retarder regulating valve body needs to be opened and closed frequently, which greatly affects the service life of the retarder regulating valve body and is prone to valve body failure. At this time, the retarder cannot work, which reduces driving safety.

[0005] Traditional hydraulic retarders have a slow exhaust process, a long time to exit operation, excessive power loss, low integration, and lack of backup solutions. If the intake solenoid valve fails to work, the hydraulic retarder cannot brake the vehicle.

[0006] In the existing retarder regulating valve body, the proportional solenoid valve is mainly used to regulate the oil intake in the retarder by adjusting the intake and exhaust. Under different air pressures, the oil intake is different, resulting in different braking torques. The proportional valve has the advantages of high adjustment accuracy and fast response. However, the proportional solenoid valve is expensive, easily damaged, resulting in a short service life and further increasing the installation and maintenance costs, making it difficult to continue to develop.

[0007] In existing, relatively mature valve body control systems, five high-speed switching solenoid valves work together, providing excellent response and high coordination accuracy. However, the operating cost remains high. Therefore, how to further reduce costs and improve the retarder adjustment system while ensuring driving safety and precise and effective adjustment are issues that need to be studied. Summary of the Invention

[0008] To address the problems existing in the prior art, this application proposes an integrated hydraulic retarder valve module adjustment and control system, which ensures driving safety while being highly integrated, reducing costs, and extending service life.

[0009] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is an integrated hydraulic retarder valve module adjustment and control system, including a hydraulic retarder valve module adjustment device and a control system. The hydraulic retarder valve module adjustment device includes an air inlet valve (2), an air source (3), a hydraulic retarder (4), a pressure relief valve (5), an exhaust valve (6), and a quick exhaust valve (7); The hydraulic retarder (4) has an air inlet for inlet and outlet of the hydraulic retarder (4) and an air outlet for exhaust gas, with the air inlet and outlet being isolated from each other. The intake end of the intake valve (2) is connected to the air source (3) through the air passage, and the exhaust end of the intake valve (2) is connected to the air inlet of the hydraulic retarder (4) through the air passage. The pressure relief valve (5), quick exhaust valve (7) and exhaust valve (6) are connected in parallel to form a parallel exhaust valve group. The air inlet of the parallel exhaust valve group is connected to the air inlet of the hydraulic retarder (4) through the air passage, and the exhaust end of the parallel exhaust valve group is connected to the atmosphere through the air outlet of the hydraulic retarder (4). The control system includes an input module, a control module, and an output module; The regulator (8) is connected to the control terminals of the intake valve (2), the quick exhaust valve (7) and the exhaust valve (6) via circuit. The regulator (8) has an MCU microprocessor and embeds a control algorithm within the MCU microprocessor. The control algorithm adopts a structure that combines multi-layer feedforward connection and feedback correction. The hidden layer neurons of the control algorithm continuously update the weight matrix through the backpropagation algorithm to accurately adjust the control parameters.

[0010] The optimized integrated hydraulic retarder valve module adjustment and control system further includes a pressure sensor (1). A pressure sensor (1) is installed in the air path near the hydraulic retarder (4) and measures and detects the pressure of the hydraulic retarder (4); the regulator (8) is connected to the control terminal of the pressure sensor (1) via a circuit.

[0011] The optimized integrated hydraulic retarder valve module adjustment control system includes CAN input, sensor input, and handle input. The CAN input section performs offset, multiplication coefficient, and filtering on the CAN signals collected from the vehicle. The CAN signals include accelerator pedal opening, brake pedal opening, vehicle speed, and ABS signal. The control module of the control system includes an initialization module, a fault diagnosis module, a graded braking control module, a constant speed braking control module, and a combined braking control module; The output modules of the control system include a solenoid valve control module, an indicator light control module, and a CAN output module.

[0012] The optimized integrated hydraulic retarder valve module regulation and control system has a three-layer connected planar module, which consists of an input layer, a hidden layer, and an output layer. The input layer has two nodes. r and y These represent system inputs respectively; The hidden layer has three nodes. It is the first neuron in the input layer that pairs with the first neuron in the hidden layer. j The weights of each neuron, It is in the hidden layer j The weights of neurons in the output layer; The output layer has one node. u As system output.

[0013] The optimized integrated hydraulic retarder valve module regulation control system described above, during the forward propagation operation of the control system, at any sampling time... k The input layer is displayed as follows: ; In the input layer, the output of each neuron corresponds to its input, which is provided by the network. and composition; This is the system's target value, corresponding to the target vehicle speed; This represents the actual value of the controlled object, corresponding to the actual vehicle speed. The three neurons in the hidden layer are proportional neurons. Integral neurons Differential neurons ; No. The input of each neuron is expressed as ; In the hidden layer, the state of each neuron is expressed as ; In the hidden layer, the output of each neuron is expressed as ; In the output layer, the input and output of a neuron are equal to the weighted sum of the outputs of neurons in the hidden layers, as shown below: ; The loss function is expressed as ; After training After several steps, the weight adjustment algorithm from the hidden layer to the output layer is expressed as follows: , in, It is the learning rate.

[0014] In the optimized integrated hydraulic retarder valve module regulation and control system, after the forward propagation calculation of the control system is completed, the backward propagation calculation begins. Consider weights For loss function The impact, i.e., calculation Based on the transfer function relationship, we obtain ; Will use replace; Characteristic function Approximate replacement sgn is the sign function; it outputs 1 when the input is positive, -1 when the input is negative, and 0 when the input is 0.

[0015] but ; The weights from the hidden layer to the output layer are adjusted to... ; After training for k steps, the weight adjustment algorithm from the hidden layer to the initial layer is expressed as follows: , in, , It is the learning rate; Derived from the weight adjustment algorithm

[0016] .

[0017] The optimized integrated hydraulic retarder valve module adjustment and control system described above includes the following steps in its parameter tuning process: 1) When At that time, assign the weights of the first neuron in the input layer to the first neuron in the hidden layer. The weights of the first neuron in the hidden layer to the neurons in the starting layer. At the same time, select the learning rate and .

[0018] 2) Under the closed-loop control architecture, multi-channel synchronous sampling technology is used to sample the reference input vector. r ( k ) and system output vector y ( k Implement time-series synchronous data acquisition and extract error signals by constructing a dynamic error assessment model. e ( k ).

[0019] 3) Based on the multi-layer topology of the PID neural network, an analytical model for signal propagation is established between the input layer, hidden layer and output layer. The algorithm is dynamically corrected through online learning.

[0020] 4) Utilize the self-learning module in the PID neural network to adjust the weights of each online layer. and ,get The adaptively adjusted parameter values.

[0021] 5) And so on, when Repeat steps 2) to 4).

[0022] In the optimized integrated hydraulic retarder valve module adjustment and control system, the intake valve (2), exhaust valve (6), and quick exhaust valve (7) are all high-speed switching solenoid valves, and the pressure relief valve (5) is a mechanical valve. The valve port diameter of the quick-release valve (7) is four to five times that of the pressure relief valve (5).

[0023] In the optimized integrated hydraulic retarder valve module adjustment and control system, the valve port diameter of the intake valve (2) is 2mm, and the valve port diameter of the quick exhaust valve (7) is 8mm.

[0024] The beneficial effects of this application are as follows: The integrated hydraulic retarder valve module regulating device provided by this invention adopts an innovative design and optimization technology that integrates electromechanical, electrical, and hydraulic systems. Based on electro-pneumatic-hydraulic combined simulation design technology, the pneumatic control system is optimized.

[0025] (2) The integrated hydraulic retarder valve module regulating device provided by this invention adopts a high-precision, fast-response integrated control system. Addressing the problems of low control accuracy, slow response speed, and high cost of electromagnetic control valves in competing products, a high-speed digitally controlled switching electromagnetic valve group is used, improving control accuracy by approximately 5% and response speed by approximately 10%. The RCU and electromagnetic valve group are highly integrated, reducing costs by 50%. Independent research and development ensures supply chain security.

[0026] (3) The integrated hydraulic retarder valve module adjustment device provided by the present invention adopts intelligent auxiliary braking control technology. Based on information fusion, working condition prediction and vehicle dynamics characteristics, a predictive intelligent vehicle speed decision system is developed, and through precise speed closed-loop control algorithm and air pressure control method, precise braking torque and vehicle speed control are achieved, effectively improving safety and intelligence level.

[0027] (4) The integrated hydraulic retarder valve module adjustment device provided by the present invention adopts high-precision melt casting technology. The stator and rotor adopt high-precision melt casting technology and an original automatic block extraction process, resulting in high product precision and good consistency, which improves production efficiency by about 50% and reduces production costs by about 70%. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the control system of this application. Figure 2 This is a schematic diagram of the framework structure of the control system of this application; Figure 3 This is a diagram of the internal architecture of the control system of this application; Figure 4 This outlines the main workflow of the parameter tuning method for the control system described in this application. Figure 5 This is a schematic diagram of the control logic of the control system of this application; Figure 6 This is a schematic diagram of the hydraulic retarder valve module regulating device of this application. Detailed Implementation

[0029] The technical features of the present invention will be further illustrated below with reference to specific embodiments.

[0030] This application provides an integrated hydraulic retarder valve module adjustment and control system, including a hydraulic retarder valve module adjustment device and a control system.

[0031] The hydraulic retarder valve module regulating device includes a pressure sensor 1, an inlet valve 2, an air source 3, a hydraulic retarder 4, a pressure relief valve 5, an exhaust valve 6, a quick exhaust valve 7, and a regulator 8. The hydraulic retarder 4 includes an inlet port and an outlet port that are isolated from each other. The inlet port is used to allow air to enter and exit the hydraulic retarder 4, and the outlet port is used to discharge exhaust gas.

[0032] The left intake end of the intake valve 2 is connected in series with the air source 3, and the right exhaust end is connected to the air inlet of the hydraulic retarder 4 through the air passage.

[0033] Pressure relief valve 5, quick exhaust valve 7 and exhaust valve 6 are connected in parallel to form a parallel exhaust valve group. The air inlet of the parallel exhaust valve group is connected to the air inlet of hydraulic retarder 4 through an air passage, and the exhaust end is connected to the atmosphere through the air outlet of hydraulic retarder 4.

[0034] Among them, intake valve 2, exhaust valve 6 and quick exhaust valve 7 are all high-speed switching solenoid valves, and pressure relief valve 5 is a mechanical valve.

[0035] The valve port diameter of quick-release valve 7 is four to five times that of exhaust valve 6. All valves are connected via air passages.

[0036] The air pressure sensor 1 is installed in the air line near the hydraulic retarder 4 to detect the air pressure of the hydraulic retarder 4.

[0037] The regulator 8 is connected to the intake valve 2, the pressure sensor 1, the quick exhaust valve 7, and the exhaust valve 6 via circuitry. The hardware connection relationship of the regulator 8 is shown in Table 1 below.

[0038] In one embodiment of this application, the valve port diameter of the intake valve 2 is 2mm, and the valve port diameter of the quick exhaust valve 7 is 8mm.

[0039] Table 1 Hardware connection relationship of regulator 8

[0040] When the hydraulic retarder 4 needs to be activated, the driver moves the retarder shift lever. The vehicle's CAN bus converts and processes the lever signals, vehicle speed signals, pedal signals, brake signals, engine speed signals, ABS signals, etc., collected by sensors, and transmits them to the control module. The control system quickly makes a judgment, finds the control air pressure corresponding to the target vehicle speed, and coordinates the high-speed switching valve of the valve module. The regulator 8 adjusts the intake valve 2 to open quickly (large duty cycle), allowing high-pressure gas from the air source 3 to enter the intake port of the hydraulic retarder 4 through the air passage. After the air pressure sensor 1 detects that the target air pressure has been reached, the regulator 8 adjusts the intake valve 2 to close. When the retarder enters the staged braking state, the vehicle's brake lights illuminate to warn following vehicles.

[0041] During operation, when the actual vehicle speed deviates from the target air pressure (i.e., the actual air pressure deviates from the target air pressure), the regulator 8 can continuously open or close the intake valve 2 and exhaust valve 6 (with a small duty cycle) through the air pressure closed-loop regulation fed back by the air pressure sensor 1, stabilizing the air pressure near the target air pressure. This, in turn, controls the actual vehicle speed near the target speed (within ±2 km / h), thus achieving closed-loop speed control. Throughout the entire operation, the air pressure sensor, water temperature sensor, and oil temperature sensor are constantly working to monitor the status of the hydraulic retarder, allowing for timely troubleshooting in case of malfunctions.

[0042] When encountering a steeper or gentler slope on a real road, i.e., when the hydraulic retarder 4 needs to increase or decrease its braking torque, it is necessary to adjust the air pressure of the hydraulic retarder 4 over a wide range. For example, when the slope becomes steeper, the air pressure needs to be adjusted from 1 bar to 2 bar. In this case, the intake valve 2 opens rapidly (large duty cycle). When the air pressure approaches 2 bar, the intake valve 2 and exhaust valve 6 are continuously opened or closed (small duty cycle) through the air pressure closed loop adjustment fed back by the air pressure sensor 1, so that the air pressure is stabilized near 2 bar. Similarly, when the slope becomes steeper, the air pressure needs to be adjusted from 2 bar to 1 bar. In this case, the exhaust valve 6 opens rapidly (large duty cycle). When the air pressure approaches 1 bar, the intake valve 2 and exhaust valve 6 are continuously opened or closed (small duty cycle) through the air pressure closed loop adjustment fed back by the air pressure sensor 1, so that the air pressure is stabilized near 1 bar. By controlling the air pressure, the actual vehicle speed is stabilized near the target vehicle speed.

[0043] When the air pressure inside the hydraulic retarder 4 exceeds the maximum air pressure (the specified safe air pressure), the pressure relief valve 5 starts to work to release air, ensuring that the air pressure is always below the maximum air pressure, thus preventing the hydraulic retarder 4 from malfunctioning and becoming dangerous.

[0044] An alarm signal is issued when the oil temperature and water temperature sensors detect temperatures exceeding the retarder's specified values, and an alarm signal is issued when the air pressure sensor detects abnormal values.

[0045] When the hydraulic retarder 4 is disengaged, the regulator 8 adjusts the exhaust valve 6 and the quick exhaust valve 7 to open quickly (100% duty cycle), so that the gas in the hydraulic retarder 4 is reversed through the air inlet of the hydraulic retarder 4 and connected to the air inlet of the exhaust valve group through the air passage, and is quickly discharged through the air outlet of the hydraulic retarder 4, so as to achieve the purpose of rapid disengagement and reduce power loss.

[0046] The integrated hydraulic retarder valve module adjustment device and its control system provided by this invention calculate and decide on the control target through its own control strategy, and output the control target to the underlying software to drive the actuator. At the same time, it uses a high-speed switching solenoid valve to adjust the hydraulic retarder, which is low in cost, durable, reliable and sensitive, and can ensure the completion of the corresponding work of the hydraulic retarder while avoiding failure. The solenoid valves are highly integrated, which facilitates the layout and installation. A mechanical pressure relief valve is used to ensure that the maximum air pressure in the retarder is less than the safe air pressure to prevent danger.

[0047] The control system is the core of the entire valve module regulating device. The control system is divided into three main modules according to its functions: the input module, the control module, and the output module.

[0048] The input module consists of three parts: CAN input, sensor input, and handle input. The CAN input section performs offsetting, multiplication, and filtering on the CAN signals acquired from the vehicle. CAN signals include accelerator pedal opening, brake pedal opening, vehicle speed, and ABS signals.

[0049] The control module includes an initialization module, a fault diagnosis module, a graded braking control module, a constant speed braking control module, and a combined braking control module.

[0050] The output modules include a solenoid valve control module, an indicator light control module, and a CAN output module.

[0051] The control system mainly controls the following parameters: Intake pressure (core control parameter): The control system controls the air pressure entering the hydraulic retarder (4) by adjusting the opening and closing of the intake valve (2) and the exhaust valve (6). The air pressure directly determines the oil pressure and oil quantity in the working chamber of the retarder, and thus determines the magnitude of the braking torque; Valve switching state and duty cycle: The control system controls the opening and closing of the high-speed switching solenoid valve via PWM (Pulse Width Modulation). Specifically, this includes: Rapid opening (high duty cycle) and closing of intake valve (2); The exhaust valve (6) is rapidly opened (high duty cycle) and closed; The intake valve (2) and exhaust valve (6) are opened and closed alternately with a small duty cycle (for fine-tuning of air pressure); The quick-release valve (7) opens rapidly at 100% duty cycle (for quick exit).

[0052] Matching target air pressure with target vehicle speed: The control system finds the corresponding target control air pressure based on the target vehicle speed, and stabilizes the actual air pressure near the target air pressure through air pressure closed-loop regulation, thereby controlling the actual vehicle speed near the target vehicle speed (between ±2km / h).

[0053] Safety protection parameters: Monitor oil temperature, water temperature, and air pressure, and issue an alarm signal when the specified values ​​are exceeded.

[0054] The CAN input signals are acquired through the vehicle's CAN bus, including accelerator pedal opening, brake pedal opening, vehicle speed, and ABS signal. The CAN input section performs offset, multiplication, and filtering on the acquired signals. Sensor input: directly acquired through hardware circuitry, including air pressure signal detected by air pressure sensor (1), water temperature signal detected by water temperature sensor, and oil temperature signal detected by oil temperature sensor. Handle input: acquired through hardware interface of the retarder gear lever signal moved by the driver.

[0055] The initialization module performs initial configuration and self-test of the control system when the system is powered on or reset.

[0056] The initialization module can be implemented in the following ways: Initialize all hardware interfaces (CAN communication, sensor acquisition, solenoid valve drive, etc.). Initialize control parameters (such as the initial weights of a PID neural network). Read system configuration parameters; Perform a power-on self-test to confirm that all hardware modules are working properly.

[0057] The fault diagnosis module is responsible for monitoring the working status of each component of the system in real time, issuing alarm signals and taking protective measures when a fault is detected.

[0058] The fault diagnosis module is implemented in the following ways: Monitor whether the detection value of the air pressure sensor (1) is abnormal, and issue an alarm signal when it is abnormal; Monitor whether the temperature detected by the oil temperature sensor and water temperature sensor exceeds the retarder's specified value, and issue an alarm signal if it does; Monitor the drive current and feedback signal of each solenoid valve to determine whether the solenoid valve is working properly; When a fault is detected, degradation protection is implemented or the output is shut down.

[0059] The function of the graded braking control module is to provide different levels of constant braking torque according to the retarder gear selected by the driver.

[0060] The functions of the graded braking control module are implemented in the following ways: Receive the driver's lever position signal to determine the target braking torque level; Find the corresponding target air pressure value based on the target braking torque; Control the intake valve (2) to open quickly and introduce high-pressure gas into the intake port of the hydraulic retarder (4); After the air pressure sensor (1) detects that the target air pressure has been reached, the intake valve (2) is closed. When the retarder enters the phased braking state, the vehicle's brake lights will illuminate to warn following vehicles.

[0061] The constant speed braking control module automatically adjusts the braking force according to the gradient when driving downhill, so that the vehicle maintains a constant speed.

[0062] The constant speed braking control module achieves its function in the following ways: Set the target vehicle speed (constant speed value); Real-time acquisition of actual vehicle speed signals; Calculate the deviation between the target vehicle speed and the actual vehicle speed; The required control air pressure is calculated using a neural network PID controller; Control the opening and closing of the intake valve (2) and exhaust valve (6) to adjust the air pressure so that the actual vehicle speed is stable near the target vehicle speed (between ±2km / h). When the slope becomes steeper, the braking torque needs to be increased (increase air pressure), the intake valve (2) opens quickly (large duty cycle), and when the air pressure is close to the target value, it switches to a small duty cycle for fine adjustment; When the slope becomes gentler, the braking torque needs to be reduced (the air pressure is reduced), the exhaust valve (6) opens quickly (large duty cycle), and when the air pressure is close to the target value, it switches to a small duty cycle for fine adjustment.

[0063] The function of the combined braking control module is to enable the hydraulic retarder to work in coordination with other braking methods such as exhaust braking and engine braking.

[0064] The combined braking control module achieves its function in the following ways: When the ABS is working, the retarder stops working; Based on braking requirements, coordinate and distribute the braking force of the hydraulic retarder with the braking force of other braking methods; Determine if the vehicle speed is less than 5 km / h; if so, turn off the exhaust brake output. Determine if there is a brake gear input; if not, shut off the exhaust brake output. Determine if the retarder is overheating; if so, activate engine braking.

[0065] The function of the solenoid valve control module is to generate and output PWM control signals to drive the opening and closing of the intake valve (2), exhaust valve (6) and quick exhaust valve (7).

[0066] The solenoid valve control module functions in the following ways: Based on the target control quantity calculated by the control module, a corresponding PWM signal is generated; Intake valve (2) control: When it is necessary to increase the air pressure, output a large duty cycle PWM signal to open it quickly; when it is necessary to fine-tune, output a small duty cycle PWM signal. Exhaust valve (6) control: When it is necessary to reduce the air pressure, output a large duty cycle PWM signal to open it quickly; when it is necessary to fine-tune, output a small duty cycle PWM signal. Quick-release valve (7) control: When the retarder exits operation, it outputs a 100% duty cycle PWM signal to quickly open; The solenoid valve coil is driven by the solenoid valve drive circuit (power amplifier circuit).

[0067] The function of the indicator light control module is to control the on / off state of indicator lights on the vehicle (such as brake lights, malfunction warning lights, etc.) and provide status prompts to the driver and vehicles behind.

[0068] How the indicator light control module works: When the retarder enters the phased braking mode, the vehicle's brake lights will illuminate to warn following vehicles. When the temperature detected by the oil temperature and water temperature sensors exceeds the retarder's specified value, the alarm indicator light will illuminate. When the barometric pressure sensor detects an abnormal value, the alarm indicator light will illuminate. The indicator light is controlled by the brake light drive circuit (switch output circuit).

[0069] The function of the CAN output module is to send the status information of the control system to other ECUs (electronic control units) in the vehicle via the CAN bus, so as to realize information sharing and collaborative control.

[0070] How the CAN output module works: Pack the retarder's operating status (such as current braking torque, operating mode, fault status, etc.) into a CAN message; Send to the vehicle's CAN bus via a CAN communication transceiver; Other ECUs in the vehicle (such as engine ECU, transmission TCU, ABS ECU, etc.) receive this information for coordinated control.

[0071] like Figure 3 As shown, the neural network PID control system adopts a design scheme that combines multi-layer feedforward connection and feedback correction. The hidden layer neurons continuously update the weight matrix through the backpropagation algorithm, thereby achieving precise adjustment of the control parameters.

[0072] Depend on Figure 3 It can be seen that the three-layer connected planar modules are respectively The neural network PID control algorithm runs in the MCU microprocessor of the regulator (8), that is, the control algorithm embedded in the MCU microprocessor in this application. Specifically, it is... Figure 3 It can be seen that the three connected planar modules are as follows: The input layer (with two nodes) uses an MCU microprocessor to acquire data via an ADC (analog-to-digital converter) interface and a CAN interface. r (Target speed) and y (Actual vehicle speed) r (Target speed) and y (Actual vehicle speed) represents the system input; Hidden layer (with three nodes): It is the first neuron in the input layer that pairs with the first neuron in the hidden layer. j The weights of each neuron, It is in the hidden layer j The weights of neurons in the output layer; Output layer (with one node): The microprocessor calculates the weighted sum of the hidden layer outputs to obtain the system output. u(Target control quantity: vehicle speed error), the microprocessor will calculate the control quantity. u The PWM duty cycle is converted into the PWM signal to drive the solenoid valve, and then output through the PWM output module. The solenoid valve performs the opening / closing action according to the PWM signal.

[0073] Forward propagation process At any sampling time k, the input to the input plane is displayed as follows: (1) In the input layer, the output of each neuron corresponds to its input, which is provided by the network. composition; It is the system's target value. This represents the actual value of the controlled object. There are three neurons in the hidden layer: Represents proportional neurons, Represents an integral neuron. Represents a differential neuron. (The...) The inputs to each neuron are as follows: (2) The state representation of each neuron in the hidden layer is as follows: (3) In the hidden layer, the output of each neuron is represented as follows: (4) In the output layer, the input and output of a neuron are equal to the weighted sum of the outputs of neurons in the hidden layers, as shown below: (5) This PID neural network controller employs the backpropagation (BP) algorithm for online adaptive training, forming a generalized network mapping of the controlled object. During training, the model continuously outputs predicted values, with mean squared error (MSE) used as the performance metric. MSE reflects model accuracy by calculating the mean squared difference between predicted and actual values; its decreasing trend directly reflects the effectiveness of training. We can represent the loss function as follows: (6) After training After these steps, the weight adjustment algorithm from the hidden layer to the output layer is shown below: (7) in, It is the learning rate.

[0074] (2) Backward propagation After the forward propagation calculation is completed, the backpropagation calculation begins. First, the weights are considered. For loss function The impact, i.e., calculation Based on the transfer function relationship, we can obtain: (8) Because the mathematical model of the controlled object is often unknown or its partial derivatives are difficult to calculate, the intermediate... The following formula is commonly used instead: (9) The characteristic function in equation (9) is used Approximate replacement Its positive or negative characteristics determine The role of the weight adjustment algorithm in the convergence direction. Here, sgn is the sign function; a positive input outputs 1, a negative input outputs -1, and a 0 input outputs 0. The advantage of using the sign function is that it normalizes to [-1, 1], avoiding large output problems caused by the denominator being very close to 0. Equation (8) can be simplified as follows: (10) (3) Weight update If we use the traditional gradient descent method, the weights from the hidden layer to the output layer can be adjusted as follows: (11) (12) After training for k steps, the weight adjustment algorithm from the hidden layer to the initial layer is shown below: (13) In the formula This is the learning rate. Similarly, the following expression can be derived: (14) (15) like Figure 4 The diagram shows the main workflow of the neural network PID controller parameter tuning method proposed in this invention.

[0075] 1) When At that time, assign the weights of the first neuron in the input layer to the first neuron in the hidden layer. The weights of the first neuron in the hidden layer to the neurons in the starting layer. At the same time, select the learning rate .

[0076] 2) Under the closed-loop control architecture, multi-channel synchronous sampling technology is used to synchronously acquire the reference input vector r(k) and the system output vector y(k) in time sequence, and the error signal e(k) is extracted by constructing a dynamic error evaluation model.

[0077] 3) Based on the multi-layer topology of the PID neural network, an analytical model for signal propagation is established between the input layer, hidden layer and output layer. The algorithm is dynamically corrected through online learning.

[0078] 4) Utilize the self-learning module in the PID neural network to adjust the weights of each online layer. and ,get The adaptively adjusted parameter values.

[0079] 5) And so on, when Repeat steps 2)-4) until the iteration terminates when any of the following conditions are met; The conditions for terminating the iteration include the error convergence condition and the maximum number of iterations condition.

[0080] Error convergence condition: The absolute value of the error signal e(k) is less than the preset allowable error threshold. (Right now This indicates that the control precision has met the requirements; the maximum number of iterations condition is: the number of iterations k reaches the preset maximum number of iterations. To prevent infinite loops.

[0081] like Figure 5 As shown, the control logic of the neural network PID is presented. Based on the simulation model of the vehicle driving downhill, the entire MATLAB / Simulink model was established.

[0082] like Figure 6 As shown, the structure of the integrated hydraulic retarder valve module adjustment device provided by the present invention specifically includes: 1. Pressure sensor, 2. Inlet valve, 3. Air source, 4. Hydraulic retarder, 5. Pressure relief valve, 6. Exhaust valve, 7. Quick exhaust valve, and 8. Regulator.

[0083] The hydraulic retarder 4 includes an air inlet and an air outlet that are isolated from each other. The air inlet is used to allow air to enter and exit the hydraulic retarder 4, and the air outlet is used to discharge exhaust gas.

[0084] The left intake end of the intake valve 2 is connected in series with the air source 3, and the right exhaust end is connected to the air inlet of the hydraulic retarder 4 through the air passage.

[0085] Pressure relief valve 5, quick exhaust valve 7 and exhaust valve 6 are connected in parallel to form a parallel exhaust valve group. The air inlet of the parallel exhaust valve group is connected to the air inlet of hydraulic retarder 4 through an air passage, and the exhaust end is connected to the atmosphere through the air outlet of hydraulic retarder 4.

[0086] Among them, intake valve 2, exhaust valve 6 and quick exhaust valve 7 are all high-speed switching solenoid valves, and pressure relief valve 5 is a mechanical valve.

[0087] The valve port diameter of quick-release valve 7 is four to five times that of exhaust valve 6.

[0088] The valves are connected by air lines.

[0089] The air pressure sensor 1 is installed in the air line near the hydraulic retarder 4 to detect the air pressure of the hydraulic retarder 4.

[0090] The regulator 8 is connected to the intake valve 2, the air pressure sensor 1, the quick exhaust valve 7 and the exhaust valve 6 via a circuit.

[0091] In another embodiment, the intake valve 2 has a valve port diameter of 2mm, and the quick exhaust valve 7 has a valve port diameter of 8mm.

[0092] When the hydraulic retarder 4 needs to be activated, the driver moves the retarder shift lever. The vehicle's CAN bus converts and processes the lever signals, vehicle speed signals, pedal signals, brake signals, engine speed signals, ABS signals, etc., collected by sensors, and transmits them to the control module. The control system quickly makes a judgment, finds the control air pressure corresponding to the target vehicle speed, and coordinates the high-speed switching valve of the valve module. The regulator 8 adjusts the intake valve 2 to open quickly (large duty cycle), allowing high-pressure gas from the air source 3 to enter the intake port of the hydraulic retarder 4 through the air passage. After the air pressure sensor 1 detects that the target air pressure has been reached, the regulator 8 adjusts the intake valve 2 to close. When the retarder enters the staged braking state, the vehicle's brake lights illuminate to warn following vehicles.

[0093] During operation, when the actual vehicle speed deviates from the target air pressure (i.e., the actual air pressure deviates from the target air pressure), the regulator 8 can continuously open or close the intake valve 2 and exhaust valve 6 (with a small duty cycle) through the air pressure closed-loop regulation fed back by the air pressure sensor 1, stabilizing the air pressure near the target air pressure. This, in turn, controls the actual vehicle speed near the target speed (within ±2 km / h), thus achieving closed-loop speed control. Throughout the entire operation, the air pressure sensor, water temperature sensor, and oil temperature sensor are constantly working to monitor the status of the hydraulic retarder, allowing for timely troubleshooting in case of malfunctions.

[0094] When encountering a steeper or gentler slope on a real road, i.e., when the hydraulic retarder 4 needs to increase or decrease its braking torque, it is necessary to adjust the air pressure of the hydraulic retarder 4 over a wide range. For example, when the slope becomes steeper, the air pressure needs to be adjusted from 1 bar to 2 bar. In this case, the intake valve 2 opens rapidly (large duty cycle). When the air pressure approaches 2 bar, the intake valve 2 and exhaust valve 6 are continuously opened or closed (small duty cycle) through the air pressure closed loop adjustment fed back by the air pressure sensor 1, so that the air pressure is stabilized near 2 bar. Similarly, when the slope becomes steeper, the air pressure needs to be adjusted from 2 bar to 1 bar. In this case, the exhaust valve 6 opens rapidly (large duty cycle). When the air pressure approaches 1 bar, the intake valve 2 and exhaust valve 6 are continuously opened or closed (small duty cycle) through the air pressure closed loop adjustment fed back by the air pressure sensor 1, so that the air pressure is stabilized near 1 bar. By controlling the air pressure, the actual vehicle speed is stabilized near the target vehicle speed.

[0095] When the air pressure inside the hydraulic retarder 4 exceeds the maximum air pressure (the specified safe air pressure), the pressure relief valve 5 starts to work to release air, ensuring that the air pressure is always below the maximum air pressure, thus preventing the hydraulic retarder 4 from malfunctioning and becoming dangerous.

[0096] An alarm signal is issued when the oil temperature and water temperature sensors detect temperatures exceeding the retarder's specified values, and an alarm signal is issued when the air pressure sensor detects abnormal values.

[0097] When the hydraulic retarder 4 is disengaged, the regulator 8 adjusts the exhaust valve 6 and the quick exhaust valve 7 to open quickly (100% duty cycle), so that the gas in the hydraulic retarder 4 is reversed through the air inlet of the hydraulic retarder 4 and connected to the air inlet of the exhaust valve group through the air passage, and is quickly discharged through the air outlet of the hydraulic retarder 4, so as to achieve the purpose of rapid disengagement and reduce power loss.

[0098] The integrated hydraulic retarder valve module adjustment device and its control system provided by this invention calculate and decide on the control target through its own control strategy, and output the control target to the underlying software to drive the actuator. At the same time, it uses a high-speed switching solenoid valve to adjust the hydraulic retarder, which is low in cost, durable, reliable and sensitive, and can ensure the completion of the corresponding work of the hydraulic retarder while avoiding failure. The solenoid valves are highly integrated, which facilitates the layout and installation. A mechanical pressure relief valve is used to ensure that the maximum air pressure in the retarder is less than the safe air pressure to prevent danger.

[0099] Of course, the above description is not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention should be protected by the present invention.

Claims

1. An integrated hydraulic retarder valve module regulation control system, characterized by: Includes hydraulic retarder valve module adjustment device and control system; The hydraulic retarder valve module adjustment device includes an air inlet valve (2), an air source (3), a hydraulic retarder (4), a pressure relief valve (5), an exhaust valve (6), and a quick exhaust valve (7); The hydraulic retarder (4) has an air inlet for inlet and outlet of the hydraulic retarder (4) and an air outlet for exhaust gas, with the air inlet and outlet being isolated from each other. The intake end of the intake valve (2) is connected to the air source (3) through the air passage, and the exhaust end of the intake valve (2) is connected to the air inlet of the hydraulic retarder (4) through the air passage. The pressure relief valve (5), quick exhaust valve (7) and exhaust valve (6) are connected in parallel to form a parallel exhaust valve group. The air inlet of the parallel exhaust valve group is connected to the air inlet of the hydraulic retarder (4) through the air passage, and the exhaust end of the parallel exhaust valve group is connected to the atmosphere through the air outlet of the hydraulic retarder (4). The control system includes a regulator (8), an input module, a control module, and an output module; The regulator (8) is connected to the control terminals of the intake valve (2), the quick exhaust valve (7) and the exhaust valve (6) via circuit. The regulator (8) has an MCU microprocessor and embeds a control algorithm within the MCU microprocessor. The control algorithm adopts a structure that combines multi-layer feedforward connection and feedback correction. The hidden layer neurons of the control algorithm continuously update the weight matrix through the backpropagation algorithm to accurately adjust the control parameters.

2. The integrated hydraulic retarder valve module regulation control system of claim 1, wherein: The hydraulic retarder valve module adjustment device also includes a pressure sensor (1). A pressure sensor (1) is installed in the air path near the hydraulic retarder (4) and measures and detects the pressure of the hydraulic retarder (4); the regulator (8) is connected to the control terminal of the pressure sensor (1) via a circuit.

3. The integrated hydraulic retarder valve module adjustment and control system according to claim 1, characterized in that: The input modules of the control system include CAN input, sensor input and handle input; the CAN input section performs offset, multiplication coefficient and filtering on the CAN signals collected from the whole vehicle; the CAN signals include accelerator pedal opening, brake pedal opening, vehicle speed and ABS signal; The control module of the control system includes an initialization module, a fault diagnosis module, a graded braking control module, a constant speed braking control module, and a combined braking control module; The output modules of the control system include a solenoid valve control module, an indicator light control module, and a CAN output module.

4. The integrated hydraulic retarder valve module regulation control system of claim 1, wherein: The control system has a three-layer interconnected planar module, which consists of an input layer, a hidden layer, and an output layer. The input layer has two nodes r and y representing the system input, respectively The hidden layer has three nodes, is the weight of the first neuron in the input layer to the j first neuron in the hidden layer, is the weight of the j neuron in the hidden layer to the neuron in the output layer; The output layer has one node, u as system output.

5. The integrated hydraulic retarder valve module adjustment and control system according to claim 4, characterized in that: The forward propagation of the control system works at any sampling time k , the input of the input layer is shown as, ; In the input layer, the output of each neuron corresponds to its input, which is input by the network and ; is the target value of the system, corresponding to the target vehicle speed; is the actual value of the controlled object, corresponding to the actual vehicle speed; The three neurons of the hidden layer are a proportional neuron , an integral neuron , a derivative neuron , respectively The input to the first neuron is expressed as ; and the input to the second neuron is expressed as ; In the hidden layer, the state of each neuron is expressed as ； In the hidden layer, the output of each neuron is expressed as ； In the output layer, the input and output of a neuron are equal to the weighted sum of the outputs of neurons in the hidden layers, as shown below: ； The loss function is expressed as ； After training After these steps, the weight adjustment algorithm from the hidden layer to the output layer is expressed as follows: ， in, It is the learning rate.

6. The integrated hydraulic retarder valve module adjustment and control system according to claim 5, characterized in that: After the forward propagation calculation of the control system is completed, the backward propagation calculation begins. Consider weights For loss function The impact, i.e., calculation Based on the transfer function relationship, we obtain ; Will use replace; Characteristic function Approximate replacement sgn is the sign function; it outputs 1 when the input is positive, -1 when the input is negative, and 0 when the input is 0. but ; The weights from the hidden layer to the output layer are adjusted to... ； After training for k steps, the weight adjustment algorithm from the hidden layer to the initial layer is expressed as follows: ， in, , It is the learning rate; Derived from the weight adjustment algorithm ； 。 7. The integrated hydraulic retarder valve module adjustment and control system according to claim 6, characterized in that: The parameter tuning process for a control system includes the following steps: 1) When At that time, assign the weights of the first neuron in the input layer to the first neuron in the hidden layer. The weights of the first neuron in the hidden layer to the neurons in the starting layer. At the same time, select the learning rate ; 2) Under the closed-loop control architecture, multi-channel synchronous sampling technology is used to synchronously acquire the reference input vector r(k) and the system output vector y(k) in time sequence, and the error signal e(k) is extracted by constructing a dynamic error evaluation model; 3) Based on the multi-layer topology of the PID neural network, an analytical model for signal propagation is established between the input layer, hidden layer and output layer. Dynamically correct through online learning algorithms; 4) Utilize the self-learning module in the PID neural network to adjust the weights of each online layer. and ,get The adaptively adjusted parameter values; 5) And so on, when Repeat steps 2)-4).

8. The integrated hydraulic retarder valve module adjustment and control system according to claim 1, characterized in that: The intake valve (2), exhaust valve (6), and quick exhaust valve (7) are all high-speed switching solenoid valves, and the pressure relief valve (5) is a mechanical valve; The valve port diameter of the quick-release valve (7) is four to five times that of the pressure relief valve (5).

9. The integrated hydraulic retarder valve module adjustment and control system according to claim 1, characterized in that: The valve port diameter of the intake valve (2) is 2mm, and the valve port diameter of the quick exhaust valve (7) is 8mm.

Images