Road surface hydraulic power generation system based on adaptive model predictive control

Abstract

The invention discloses a pavement hydraulic power generation system based on adaptive model predictive control. The pavement hydraulic power generation system comprises a pavement hydraulic power generation device and a control system. The road surface hydraulic power generation device comprises a hydraulic cylinder, a hydraulic control reversing valve, a single-action pressure cylinder, a speed regulating valve, a one-way valve, an energy accumulator, an electromagnetic reversing valve, a throttling valve, an overflow valve, a hydraulic motor and a power generator. A piston rod of the hydraulic cylinder is connected with a road surface speed bump, an oil outlet of the hydraulic cylinder is connected with a left cavity of the single-action pressure cylinder through the hydraulic control reversing valve, a right cavity of the single-action pressure cylinder is connected with an oil inlet of the energy accumulator through the speed regulating valve and the one-way valve, and an oil outlet of the energy accumulator is connected with an oil inlet of the hydraulic motor through the electromagnetic reversing valve and the throttling valve. The driving end of the hydraulic motor is connected with the generator. The overflow valve is connected with the hydraulic motor in parallel. The control system takes an AMPC algorithm as a decision center to form closed-loop control of'sensing-prediction-optimization-execution-feedback ', so that the power generation system can quickly adapt to complex road conditions such as road surface pressure change and traffic flow change.

Description

Technical Field

[0001] This invention belongs to the field of road pressure energy power generation technology, specifically a road hydraulic power generation system based on adaptive model predictive control. Background Technology

[0002] Road pressure power generation technology converts the pressure energy generated by vehicles rolling over the road surface into electrical energy, providing a new energy pathway for powering and storing road infrastructure, thus achieving efficient energy utilization and sustainable environmental development. However, road pressure power generation systems face complex and variable operating conditions in actual operation, such as large fluctuations in road pressure and unstable traffic density. These factors seriously affect energy conversion efficiency and system operational stability.

[0003] Traditional pavement pressure power generation systems mostly employ conventional PID control or simple logic control strategies, which are ill-suited to adapting to the dynamic changes of complex operating conditions. Furthermore, they fail to meet practical application requirements in terms of energy conversion efficiency, equipment loss control, and operational stability, thus limiting the promotion and development of pavement pressure power generation technology. Therefore, developing a power generation system that can adapt to complex operating conditions, balance multiple objectives, and possess high reliability is crucial for promoting the practical application of pavement pressure power generation technology. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the technical problem this invention aims to solve is to provide a road hydraulic power generation system based on adaptive model predictive control.

[0005] The present invention solves the aforementioned technical problem by adopting the following technical solution:

[0006] A road hydraulic power generation system based on adaptive model predictive control, characterized in that it includes a road hydraulic power generation device and a control system;

[0007] The road hydraulic power generation device includes a hydraulic cylinder, a hydraulically controlled directional valve, a single-acting booster cylinder, a speed control valve, a check valve, an accumulator, a solenoid directional valve, a throttle valve, a relief valve, a hydraulic motor, and a generator. The piston rod of the hydraulic cylinder is connected to the road speed bump. The oil outlet of the hydraulic cylinder is connected to the left chamber of the single-acting booster cylinder via the hydraulically controlled directional valve. The right chamber of the single-acting booster cylinder is connected to the oil inlet of the accumulator via the speed control valve and the check valve. The oil outlet of the accumulator is connected to the oil inlet of the hydraulic motor via the solenoid directional valve and the throttle valve. The drive end of the hydraulic motor is connected to the generator. The relief valve is connected in parallel with the hydraulic motor.

[0008] The control system includes an intelligent sensing and state monitoring layer, an adaptive model predictive control layer, a precise execution and collaborative control layer, and a feedback optimization and fault diagnosis layer.

[0009] In the adaptive model predictive control layer, firstly, the system state-space equations are constructed:

[0010]

[0011] In the formula, Let x(t) represent the rate of change of the system state variable, u(t) represent the throttle valve control current, w represent the disturbance term, y(t) represent the generator power, v represent the measurement noise, and A, B, and C represent the parameter matrices.

[0012] Then, based on the multi-objective function of formula (2) which has the highest power generation efficiency, stable energy storage and minimum energy consumption, the optimal control current of the throttle valve is obtained by solving the problem;

[0013] J=ω1J1+ω2J2+ω3J3 (2)

[0014] in, For the deviation in power generation efficiency, P shiji P zuiyou These are the actual power generation efficiency and the optimal power generation efficiency, respectively. Let P0 be the initial pressure of the accumulator, and P(t) be the pressure of the accumulator at time t. The values ​​are energy consumption indicators, where Q is the hydraulic oil flow rate at the inlet of the hydraulic motor, and P is the hydraulic oil pressure at the inlet of the hydraulic motor. ω1, ω2, and ω3 are the electrical power output by the generator, and ω1, ω2, and ω3 are all weighting coefficients.

[0015] Finally, the optimal control current of the throttle valve is substituted into the system state-space equation to obtain the predicted system output value. The actual system output value is compared with the predicted value, and the parameter matrix of the system state-space equation is updated by recursive least squares method to obtain the optimal control current of the throttle valve again, thus realizing the closed-loop control of the throttle valve opening.

[0016] Furthermore, the feedback optimization and fault diagnosis layer calculates the throttle valve opening deviation based on the actual and expected opening degrees of the throttle valve; when the throttle valve opening deviation exceeds a deviation threshold, a correction amount is calculated. Where e(t) is the throttle valve opening deviation, It is the first derivative of the throttle valve opening deviation, K p K i and K d These are the proportional coefficient, integral coefficient, and derivative coefficient, respectively; the adaptive model predictive control layer adjusts the optimal control current of the throttle valve based on the correction amount.

[0017] Fault diagnosis is performed using a fault diagnosis model built on a neural network, based on multimodal sensing data collected by the intelligent sensing and state monitoring layer.

[0018] Furthermore, the intelligent sensing and state monitoring layer acquires multimodal sensing data through multimodal sensors, including pressure sensors, flow sensors, temperature sensors, and vibration sensors. The pressure sensors are located at the oil outlet of the accumulator, the inlet and outlet of the throttle valve, and the oil inlet of the hydraulic motor, respectively. The flow sensor is located at the oil inlet of the hydraulic motor, the temperature sensor is located on the housing of the hydraulic motor, and the vibration sensor is located on the bearing seat of the hydraulic motor.

[0019] Compared with the prior art, the beneficial effects of the present invention are:

[0020] The power generation device of this invention is based on an energy harvesting mechanism, a pressurization mechanism, an energy storage mechanism, and an energy conversion mechanism. It collects the pressure energy generated by vehicles passing by on the road, converts it into the pressure energy of hydraulic oil, further converts it into the pressure energy of an accumulator, and finally converts it into electrical energy through a generator. The control system revolves around the core link of accumulator-throttle valve-hydraulic motor-generator, using an adaptive model predictive control algorithm as the decision-making center to construct a closed-loop system of "perception-prediction-optimization-execution-feedback". The intelligent perception and state monitoring layer collects system state variables in real time. The AMPC algorithm combines the system state space equation to predict engine efficiency, optimizes the throttle valve opening, and generates control commands. The precise execution and collaborative control layer precisely adjusts the throttle valve opening according to the control commands. The feedback optimization and fault diagnosis layer monitors the actual throttle valve opening and generates a correction amount based on the throttle valve opening deviation. The adaptive model predictive control layer adjusts the control commands according to the correction amount, realizing full-process optimization from data perception and dynamic prediction to precise control.

[0021] Through adaptive updates of the system's state space, the system can quickly adapt to complex road conditions such as road pressure fluctuations and traffic flow changes. Based on a multi-objective optimization control strategy, it effectively balances power generation efficiency, energy storage stability, and energy consumption control. Combined with fault diagnosis and feedback adjustment mechanisms, the system's reliability and stability are improved. Compared to traditional control systems, this power generation system significantly improves energy conversion efficiency and reduces equipment losses, providing an efficient and intelligent solution for the practical application of road pressure power generation technology. Attached Figure Description

[0022] Figure 1 This is a structural diagram of the power generation system;

[0023] Figure 2 This is the overall flowchart of the control system;

[0024] Figure 3 A flowchart for the intelligent sensing and state monitoring layer;

[0025] Figure 4 A flowchart for predicting the control layer in an adaptive model;

[0026] Figure 5 The flowchart for precise execution and collaborative control layer;

[0027] Figure 6 A flowchart for the feedback optimization and fault diagnosis layer;

[0028] Among them, 1-hydraulic cylinder; 2-hydraulic directional valve; 3-single-acting booster cylinder; 4-speed control valve; 5-check valve; 6-accumulator; 7-solenoid directional valve; 8-throttle valve; 9-relief valve; 10-hydraulic motor; 11-generator; 12-oil tank. Detailed Implementation

[0029] Specific embodiments are given below with reference to the accompanying drawings. These specific embodiments are only used to describe the technical solution of the present invention in detail, and are not intended to limit the scope of protection of this application.

[0030] This invention provides a road hydraulic power generation system based on adaptive model predictive control, comprising two parts: a road hydraulic power generation device and a control system;

[0031] The road hydraulic power generation device includes a hydraulic cylinder 1, a hydraulically controlled directional valve 2, a single-acting booster cylinder 3, a speed regulating valve 4, a one-way valve 5, an accumulator 6, a solenoid directional valve 7, a throttle valve 8, an overflow valve 9, a hydraulic motor 10, and a generator 11.

[0032] Hydraulic cylinder 1 is installed at the speed bump on the highway. The piston rod of hydraulic cylinder 1 is fixed to the speed bump. When a vehicle runs over the speed bump, it pushes the piston rod of hydraulic cylinder 1 downward, causing hydraulic cylinder 1 to output hydraulic oil. The oil outlet of hydraulic cylinder 1 is connected to the left chamber of single-acting booster cylinder 3 through hydraulic control directional valve 2. The right chamber of single-acting booster cylinder 3 is connected to the oil inlet of accumulator 6 through speed control valve 4 and check valve 5. The oil outlet of accumulator 6 is connected to the oil inlet of hydraulic motor 10 through solenoid directional valve 7 and throttle valve 8. The drive end of hydraulic motor 10 is connected to generator 11. Overflow valve 9 is connected in parallel with hydraulic motor 10. Single-acting booster cylinder 3, hydraulic motor 10 and overflow valve 9 are connected to oil tank 12.

[0033] Hydraulic cylinder 1 serves as the energy harvesting mechanism, single-acting booster cylinder 3 as the boosting mechanism, accumulator 6 as the energy storage mechanism, and hydraulic motor 10 and generator 11 as the energy conversion mechanism. Initially, the valve core of hydraulic directional valve 2 is in the left position, and solenoid directional valve 7 is in the right position, with the system in a fluid-filled state. When a vehicle rolls over a speed bump, the piston rod of hydraulic cylinder 1 moves downward, pushing hydraulic oil from the rodless chamber of hydraulic cylinder 1 into the left chamber of single-acting booster cylinder 3. The hydraulic oil pressure increases after passing through single-acting booster cylinder 3, then enters accumulator 6 through speed control valve 4 and check valve 5, thereby converting road pressure energy into hydraulic oil pressure energy and storing it in accumulator 6, thus realizing the system's energy conversion. Oil intake; if the vehicle's weight is large, resulting in high hydraulic oil pressure, the hydraulic control directional valve 2 will move the valve core under the action of oil pressure, placing it in the right position. The hydraulic oil in the hydraulic cylinder 1 cannot enter the single-acting booster cylinder 3 for boosting, thus maintaining the system pressure and avoiding system shock and instability caused by boosting again when the hydraulic oil pressure is too high. When the vehicle leaves the speed bump, the piston rod of the hydraulic cylinder 1 returns to the initial position under the action of the spring. The hydraulic oil in the left chamber of the single-acting booster cylinder 3 flows back to the rodless chamber of the hydraulic cylinder 1, and the right chamber of the single-acting booster cylinder 3 draws hydraulic oil from the oil tank 12, causing the piston of the single-acting booster cylinder 3 to return to the initial position, realizing system oil return.

[0034] When the hydraulic oil in the accumulator 6 reaches a certain amount, it is released. The hydraulic oil enters the hydraulic motor 10 through the solenoid directional valve 7 and the throttle valve 8, driving the hydraulic motor 10 to rotate. The hydraulic motor 10 drives the generator 11 to work, realizing the generation of electricity from road pressure energy.

[0035] Because the minimum operating pressure of the accumulator needs to be set within a relatively high range, and when a light-weight vehicle rolls over the road surface, the circuit pressure is too low to store energy, a single-acting booster cylinder 3 is introduced to ensure the efficiency of the accumulator. This cylinder increases the pressure several times over when the input pressure is low, improving energy utilization and system efficiency. The generator combines the single-acting booster cylinder 3 and the hydraulic control directional valve 2, which can automatically adjust the operating mode according to the vehicle weight. One is the normal operating mode, where the hydraulic cylinder 1 receives a suitable rolling pressure, the hydraulic control directional valve 2 is in the left position, and the hydraulic oil in the hydraulic cylinder 1 enters the single-acting booster cylinder 3 for normal pressurization. The other is the overpressure protection mode, where the hydraulic cylinder 1 receives excessive rolling pressure, the hydraulic control directional valve 2 switches to the right position due to pressure, preventing the hydraulic oil in the hydraulic cylinder 1 from entering the single-acting booster cylinder 3, thus preventing the high-pressure hydraulic oil from being pressurized again and protecting the circuit. This improves the energy storage efficiency under light load conditions and protects the circuit under heavy load conditions, preventing damage to the system from excessive pressure. Compared to traditional hydraulic energy storage systems, this design offers greater adaptability and efficiency, better meeting the energy recovery and utilization needs under different operating conditions.

[0036] like Figure 2As shown, the control system adopts a hierarchical architecture, including an intelligent sensing and state monitoring layer, an adaptive model predictive control layer, a precise execution and collaborative control layer, and a feedback optimization and fault diagnosis layer. Real-time data interaction is achieved between these layers via a data bus. The intelligent sensing and state monitoring layer monitors system state variables, while the adaptive model predictive control layer obtains predicted system output values ​​based on these variables, generates control commands, calculates prediction deviations, and corrects the control commands based on these deviations.

[0037] like Figure 3 As shown, the intelligent sensing and state monitoring layer acquires multimodal sensing data reflecting the system's operating state, including hydraulic oil pressure, hydraulic oil flow rate, temperature, and vibration, through multimodal sensors, and processes the data to generate key feature vectors. The multimodal sensors include pressure sensors, flow sensors, temperature sensors, and vibration sensors. Pressure sensors are installed at the oil outlet of the accumulator 6, the inlet and outlet of the throttle valve 8, and the oil inlet of the hydraulic motor 10 to measure hydraulic oil pressure in real time and obtain energy transmission status by sensing changes in hydraulic oil pressure. A flow sensor is installed at the oil inlet of the hydraulic motor 10 to measure hydraulic oil pressure in real time. Temperature sensors are installed at the inlet of the oil tank 12 and on the housing of the hydraulic motor 10 to measure the temperature in real time. A vibration sensor is installed on the bearing housing of the hydraulic motor 10 to measure the vibration signal of the hydraulic motor 10 in real time and reflect the wear level through the vibration state. The signals of the multimodal sensors are amplified, filtered and converted from analog to digital to obtain multimodal sensing data. The multimodal sensing data is normalized and then the principal component analysis (PCA) algorithm is used to reduce the dimensionality of the normalized multimodal sensing data. The data is then transmitted to the next layer through the SPI interface (transmission rate 10Mbps) to provide an accurate and refined data foundation for subsequent decision-making.

[0038] like Figure 4 As shown, the adaptive model predictive control layer is based on an improved adaptive model predictive control algorithm to form a complete closed-loop control; first, the system state-space equations are constructed based on physical laws;

[0039]

[0040] In the formula, The system state variable x(t) represents the rate of change of the system state variable, used to describe the dynamic changes of the system state over time. The system state variable x(t) includes accumulator pressure (unit: MPa), hydraulic motor speed (unit: r / min), and throttle valve opening (unit: %). These variables comprehensively reflect the key operating states of the system during the energy conversion process. u(t) represents the control variable, which is the throttle valve control current (unit: mA). By adjusting this control current, the throttle valve opening is controlled, thereby adjusting parameters such as system pressure and flow rate to achieve system control. w represents the disturbance term, representing external interference factors of the system, including road pressure fluctuations, oil temperature changes, etc. y(t) represents the output variable, which is the generator power. v represents the measurement noise, which is an unavoidable error in the output measurement process. A, B, and C represent parameter matrices.

[0041] A multi-objective function is constructed based on maximizing power generation efficiency, ensuring stable energy storage, and minimizing energy consumption. The particle swarm optimization algorithm is used to solve the multi-objective function to obtain the optimal control command, i.e., the optimal control current for the throttle valve. The multi-objective function is:

[0042] J=ω1J1+ω2J2+ω3J3(2)

[0043] in, For the deviation in power generation efficiency, P shiji P zuiyou These are the actual power generation efficiency and the optimal power generation efficiency, respectively. Let P0 be the initial pressure of the accumulator, and P(t) be the pressure of the accumulator at time t. The values ​​are energy consumption indicators, where Q is the hydraulic oil flow rate at the inlet of the hydraulic motor, and P is the hydraulic oil pressure at the inlet of the hydraulic motor. ω1, ω2, and ω3 are the electrical power output by the generator, and ω1, ω2, and ω3 are all weighting coefficients.

[0044] Substituting the optimal control current of the throttle valve into the system state-space equation yields the predicted system output value. The actual system output value is compared with the predicted value, and the parameter matrix of the system state-space equation is updated using the recursive least squares method. The forgetting factor is set to 0.95-0.99. When the prediction deviation (the difference between the actual system output value and the predicted value) is greater than the set value, a forced update is triggered. The control command is regenerated based on the updated system state-space equation to achieve closed-loop optimization control. The rolling update cycle is synchronized with the control cycle (10ms) to ensure that the prediction deviation is less than or equal to the set value, thereby ensuring stable system operation.

[0045] like Figure 5As shown, the precise execution and collaborative control layer is responsible for converting the optimal control command generated by the adaptive model predictive control layer into the throttle valve opening degree to control the hydraulic oil flow rate, thereby adjusting the energy transmission rate and achieving efficient conversion of hydraulic energy into electrical energy. First, the optimal control command is parsed, that is, the optimal control command is converted into the throttle valve control current (4-20mA corresponds to an opening degree of 0-100%) by a lookup table method, with a parsing delay of ≤5ms and a conversion accuracy of ±0.5%FS. The actual opening degree of the throttle valve is collected in real time and transmitted to the feedback optimization and fault diagnosis layer, with a feedback cycle of 10ms, to ensure that the execution deviation is less than or equal to the set value (≤2%).

[0046] like Figure 6 As shown, the feedback optimization and fault diagnosis layer continuously optimizes the control strategy through closed-loop feedback and ensures the safe and reliable operation of the system through real-time fault diagnosis. First, the throttle valve opening deviation e(t) is calculated based on the actual opening degree and the expected opening degree of the throttle valve. When the throttle valve opening deviation is greater than the set deviation threshold e... th When the error reaches 5%, local optimization is initiated, and a PID algorithm (proportional coefficient 0.5-2.0, integral time 0.1-1.0s, derivative time 0-0.5s) is used to calculate the correction. Among them, K p K i and K d These are the proportional coefficient, integral coefficient, and differential coefficient, respectively.

[0047] It is the first derivative of the throttle valve opening deviation; the correction is transmitted to the adaptive model predictive control layer, which adjusts the control command accordingly to achieve feedback optimization of the control command. The optimization cycle is synchronized with the control cycle to ensure that the throttle valve opening deviation after closed-loop regulation is less than or equal to the set deviation threshold. Then, the system pressure, flow rate and temperature are collected in real time and input into the fault diagnosis model to achieve fault diagnosis. The fault diagnosis model is built based on a BP neural network and trained using historical data. It can also monitor pressure surges (≥10.5MPa / s), abnormal flow (deviation from set value ±20%), temperature exceeding limits (≥80℃), and vibration exceeding limits (effective value ≥5g) in real time. When a first-level fault (minor deviation) occurs, the fault is eliminated through parameter self-adjustment; when a second-level fault occurs, the control strategy is adjusted (response time ≤200ms); when a third-level fault (serious fault) occurs, safety protection is activated, the hydraulic oil circuit is cut off (execution delay ≤500ms) and an alarm signal is issued (audible and visual alarm + RS485 communication alarm).

[0048] Matters not covered in this invention are common knowledge.

Claims

1. A road hydraulic power generation system based on adaptive model predictive control, characterized in that, This includes road surface hydraulic power generation devices and control systems; The road hydraulic power generation device includes a hydraulic cylinder, a hydraulically controlled directional valve, a single-acting booster cylinder, a speed control valve, a check valve, an accumulator, a solenoid directional valve, a throttle valve, a relief valve, a hydraulic motor, and a generator. The piston rod of the hydraulic cylinder is connected to the road speed bump. The oil outlet of the hydraulic cylinder is connected to the left chamber of the single-acting booster cylinder via the hydraulically controlled directional valve. The right chamber of the single-acting booster cylinder is connected to the oil inlet of the accumulator via the speed control valve and the check valve. The oil outlet of the accumulator is connected to the oil inlet of the hydraulic motor via the solenoid directional valve and the throttle valve. The drive end of the hydraulic motor is connected to the generator. The relief valve is connected in parallel with the hydraulic motor. The control system includes an intelligent sensing and state monitoring layer, an adaptive model predictive control layer, a precise execution and collaborative control layer, and a feedback optimization and fault diagnosis layer. In the adaptive model predictive control layer, firstly, the system state-space equations are constructed: In the formula, Let x(t) represent the rate of change of the system state variable, u(t) represent the throttle valve control current, w represent the disturbance term, y(t) represent the generator power, v represent the measurement noise, and A, B, and C represent the parameter matrices. Then, based on the multi-objective function of formula (2) which has the highest power generation efficiency, stable energy storage and minimum energy consumption, the optimal control current of the throttle valve is obtained by solving the problem; J=ω1J1+ω2J2+ω3J3 (2) in, For the deviation in power generation efficiency, P shiji P zuiyou These are the actual power generation efficiency and the optimal power generation efficiency, respectively. Let P0 be the initial pressure of the accumulator, and P(t) be the pressure of the accumulator at time t. The values ​​are energy consumption indicators, where Q is the hydraulic oil flow rate at the inlet of the hydraulic motor, and P is the hydraulic oil pressure at the inlet of the hydraulic motor. ω1, ω2, and ω3 are the electrical power output by the generator, and ω1, ω2, and ω3 are all weighting coefficients. Finally, the optimal control current of the throttle valve is substituted into the system state-space equation to obtain the predicted system output value. The actual system output value is compared with the predicted value, and the parameter matrix of the system state-space equation is updated by recursive least squares method to obtain the optimal control current of the throttle valve again, thus realizing the closed-loop control of the throttle valve opening.

2. The road hydraulic power generation system based on adaptive model predictive control according to claim 1, characterized in that, The feedback optimization and fault diagnosis layer calculates the throttle valve opening deviation based on the actual and expected opening degrees; when the throttle valve opening deviation exceeds a deviation threshold, a correction amount is calculated. Where e(t) is the throttle valve opening deviation, K is the first derivative of the throttle valve opening deviation. p K i and K d These are the proportional coefficient, integral coefficient, and derivative coefficient, respectively; the adaptive model predictive control layer adjusts the optimal control current of the throttle valve based on the correction amount. Fault diagnosis is performed using a fault diagnosis model built on a neural network, based on multimodal sensing data collected by the intelligent sensing and condition monitoring layer.

3. The road hydraulic power generation system based on adaptive model predictive control according to claim 1 or 2, characterized in that, The intelligent sensing and status monitoring layer acquires multimodal sensing data through multimodal sensors, including pressure sensors, flow sensors, temperature sensors, and vibration sensors. The pressure sensors are located at the oil outlet of the accumulator, the inlet and outlet of the throttle valve, and the oil inlet of the hydraulic motor, respectively. The flow sensor is located at the oil inlet of the hydraulic motor, the temperature sensor is located on the housing of the hydraulic motor, and the vibration sensor is located on the bearing seat of the hydraulic motor.

Images