Wave energy power generation simulation demonstrator with adaptive algorithm

Abstract

The application discloses a wave energy power generation simulation demonstrator with an adaptive algorithm, which comprises a bearing base, a water tank fixed on the base, a wave generation module, an energy capture module, a power generation module, a data display module and an edge computing intelligent regulation and control module; the driving part of the wave generation module is arranged outside the water tank, and the execution part extends into the water tank to generate waves; the energy capture module comprises a float floating on the water surface and converts wave energy into mechanical energy; the power generation module converts mechanical energy into electrical energy; the data display module collects and displays power generation parameters; the edge computing intelligent regulation and control module is internally provided with an adaptive algorithm, obtains real-time power generation data from the data display module, dynamically optimizes wave generation parameters of the wave generation module and energy capture parameters of the energy capture module, and realizes intelligent calibration of wave working conditions and optimal matching of power generation efficiency. The application has compact structure, high intelligent degree, can accurately simulate real marine wave working conditions and adaptively optimize power generation efficiency.

Description

Technical Field

[0001] This invention relates to the field of new energy teaching experimental devices and intelligent simulation technology, and in particular to a wave energy power generation simulator equipped with an adaptive algorithm. Background Technology

[0002] Currently, most wave energy power generation teaching demonstration devices on the market adopt fixed mechanical structures and simple electrical control schemes. Typically, a motor with a fixed rotation speed drives a pusher plate to generate regular waves, and a float drives a micro generator through a connecting rod to generate electricity, which is displayed by a light bulb or a simple voltmeter. These devices are simple in structure and low in cost, and are widely used in new energy science popularization teaching scenarios in primary and secondary schools.

[0003] However, existing technologies have significant limitations. Traditional equipment can only generate regular waves with fixed frequencies and heights, failing to simulate complex conditions such as random and superimposed waves in the real ocean, resulting in extremely low simulation fidelity. The equipment lacks intelligent control capabilities; all parameters are manually set and cannot be dynamically adjusted based on power generation efficiency, making autonomous optimization and parameter iteration impossible during experiments. Furthermore, traditional teaching aids can only perform qualitative demonstrations, lacking high-precision data acquisition and quantitative analysis functions, thus failing to support rigorous control experiments and scientific research. These technological bottlenecks lead to serious deficiencies in the depth of teaching, experimental precision, and simulation realism of existing equipment.

[0004] Therefore, there is an urgent need to develop a wave energy power generation simulator that can simulate complex ocean wave conditions, has intelligent adaptive control capabilities, and supports high-precision data acquisition and quantitative analysis, in order to meet the urgent demand for high-end simulation experimental equipment in the field of new energy teaching and research. Summary of the Invention

[0005] To improve existing methods and systems, a wave energy generation simulator with an adaptive algorithm is provided, including: Support base; The water tank is fixedly installed on the support base; The wave generation module has its drive unit located outside the water tank and its execution unit extending inside the water tank. The wave generation module is configured to generate waves in the water within the water tank. The energy capture module includes a float that can float on the surface of the water in the tank. The energy capture module is configured to capture wave energy and convert the wave energy into mechanical energy. The power generation module, connected to the energy capture module, is configured to convert mechanical energy into electrical energy; The data display module, electrically connected to the power generation module, is configured to collect and display power generation parameters; The edge computing intelligent control module is electrically connected to the data display module, wave generation module, and energy capture module, respectively. Among them, the edge computing intelligent control module has a built-in adaptive algorithm, which is configured to obtain real-time power generation data from the data display module, and dynamically optimize the wave generation parameters of the wave generation module and the energy capture parameters of the energy capture module based on the real-time power generation data, so as to achieve intelligent calibration of wave conditions and optimal matching of power generation efficiency.

[0006] In some embodiments, the adaptive algorithm is a lightweight deep Q-network reinforcement learning algorithm, configured to collect power generation, wave frequency and wave height data in real time, and dynamically and adaptively adjust the motor speed and pusher amplitude of the wave generation module, as well as the buoy depth of the energy capture module through algorithm iterative calculation.

[0007] In some embodiments, the edge computing intelligent control module also incorporates a wave spectrum fitting algorithm, which is configured to simulate random waves, clutter, and / or superimposed wave conditions in the real ocean.

[0008] In some embodiments, the edge computing intelligent control module is also configured to support switching between manual mode and intelligent adaptive mode; In manual mode, users can manually adjust the motor speed of the wave generation module, the hinge position of the connecting rod on the eccentric wheel, and the water depth of the float of the energy capture module. In intelligent adaptive mode, the edge computing intelligent control module autonomously adjusts the parameters of the wave generation module and the energy capture module based on the adaptive algorithm.

[0009] In some embodiments, the energy harvesting module further includes a support, a vertical guide slide, and a crank-connecting rod transmission mechanism; The support frame is installed above the water tank; At least two vertical guide slide rods are fixed vertically to the bracket, and the float is slidably sleeved on the outside of the vertical guide slide rods and can float up and down along the vertical guide slide rods; The upper end of the float is connected to a crank-connecting rod transmission mechanism, which is configured to convert the linear reciprocating motion of the float into circular rotational motion and transmit it to the power generation module.

[0010] In some embodiments, the vertical guide slide bar is provided with a plurality of positioning buckles, which are configured to adjust the fixed position of the float on the vertical guide slide bar, thereby adjusting the water depth of the float.

[0011] In some embodiments, the adaptive algorithm is also configured to autonomously iteratively optimize the water depth parameters of the buoy based on real-time power generation data.

[0012] In some embodiments, the wave generation module includes a speed-regulating drive motor, an eccentric wheel, a connecting rod, and a oscillating pusher plate; The speed-regulating drive motor is fixedly installed on the outer wall of the water tank, and the output shaft of the speed-regulating drive motor is fixedly connected to the eccentric wheel. The eccentric wheel is provided with several radially distributed hinge holes. One end of the connecting rod is detachably hinged to one of the hinge holes, and the other end of the connecting rod is hinged to the swing push plate. The oscillating pusher plate is vertically installed inside the water tank; The wave height can be adjusted by selecting different hinge holes and adjusting the swing amplitude of the oscillating pusher plate.

[0013] In some embodiments, the data display module includes a voltage sensor, a current sensor, and a liquid crystal digital display screen; The voltage sensor and the current sensor are connected in series to the output circuit of the power generation module. The signal output terminals of the voltage sensor and the current sensor are both connected to the LCD digital display screen and are configured to display the output voltage and current values ​​in real time. The voltage and current sensors are also connected to the edge computing intelligent control module and are configured to provide real-time power generation data to the edge computing intelligent control module.

[0014] In some embodiments, the support base is a hollow box structure, and the interior of the support base houses electrical connection lines and an edge computing intelligent control module; A main power switch is installed on the outside of the support base, which is configured to control the overall power-on and power-off of the wave generation module, data display module, and edge computing intelligent control module.

[0015] This invention proposes a wave energy generation simulator with an adaptive algorithm, comprising a support base, a water tank, a wave generation module, an energy capture module, a power generation module, a data display module, and an edge computing intelligent control module. The drive part of the wave generation module is located outside the water tank, while the execution part extends into the water tank to generate waves. The float of the energy capture module floats on the water surface to capture wave energy and convert it into mechanical energy, which is then converted into electrical energy by the power generation module. The data display module collects and displays power generation parameters. The edge computing intelligent control module has a built-in adaptive algorithm that obtains real-time power generation data from the data display module and dynamically optimizes the wave generation parameters of the wave generation module and the energy capture parameters of the energy capture module. The beneficial effects of this invention are: by using a built-in adaptive algorithm to achieve intelligent calibration of wave conditions and optimal matching of power generation efficiency, it can accurately simulate complex wave conditions in the real ocean, possesses autonomous learning and adaptive optimization capabilities, supports high-precision data acquisition and quantitative analysis, and significantly improves the simulation, scientific rigor, and intelligence level of teaching experiments. Attached Figure Description

[0016] Figure 1This is a system architecture diagram of the wave energy power generation simulator with adaptive algorithm proposed in this invention.

[0017] In the diagram: 1. Support base; 2. Water tank; 3. Wave generation module; 4. Energy capture module; 5. Power generation module; 6. Data display module; 7. Edge computing intelligent control module. Detailed Implementation

[0018] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0019] Firstly, this application provides a wave energy generation simulator with an adaptive algorithm, such as... Figure 1 As shown, it includes: Support base 1; Water tank 2 is fixedly installed on the support base 1; Wave generation module 3, the drive part of wave generation module 3 is located outside water tank 2, the execution part of wave generation module 3 extends into water tank 2, and wave generation module 3 is configured to generate waves in the water body inside water tank 2. The energy capture module 4 includes a float that can float on the surface of the water in the water tank 2. The energy capture module 4 is configured to capture wave energy and convert the wave energy into mechanical energy. The power generation module 5, connected to the energy capture module 4, is configured to convert mechanical energy into electrical energy; Data display module 6, electrically connected to power generation module 5, is configured to collect and display power generation parameters; The edge computing intelligent control module 7 is electrically connected to the data display module 6, the wave generation module 3, and the energy capture module 4, respectively. Among them, the edge computing intelligent control module 7 has a built-in adaptive algorithm, which is configured to obtain real-time power generation data from the data display module 6, and dynamically optimize the wave generation parameters of the wave generation module 3 and the energy capture parameters of the energy capture module 4 based on the real-time power generation data, so as to achieve intelligent calibration of wave conditions and optimal matching of power generation efficiency.

[0020] This embodiment provides a wave energy power generation simulator equipped with an adaptive algorithm, the overall structure of which is as follows: Figure 1 As shown. This demonstrator is mainly used for teaching experiments on new energy sources, and it can simulate real ocean wave conditions and achieve adaptive optimization of power generation efficiency.

[0021] The support base 1 serves as the foundation for the entire device. It employs a hollow box structure and can be made of engineering plastic or aluminum alloy. The support base 1 not only provides a stable mounting surface for the modules above, but its internal cavity also houses the electrical connection lines and the edge computing intelligent control module 7, preventing exposed wiring and improving the device's safety and tidiness. A main power switch is located on the outside of the support base 1 to control the power supply and shutdown of the entire device, allowing for convenient one-button operation by the user.

[0022] Water tank 2 is fixedly installed on the supporting base 1 and is made of high-transparency acrylic sheet, facilitating observation of internal water fluctuations and mechanical transmission processes. Millimeter-scale markings are engraved on the side walls of water tank 2 for real-time wave height readings, providing raw physical parameters for subsequent algorithm calibration. The dimensions of water tank 2 can be flexibly designed according to teaching needs; for example, it can be 40 cm long, 25 cm wide, and 30 cm high, sufficient to hold enough experimental water and ensure adequate space for wave propagation.

[0023] The wave generation module 3 is used to generate controllable waves in the water within the water tank 2. Its drive unit is located outside the water tank 2, while its execution unit extends into the water tank 2. In a preferred embodiment, the drive unit includes a variable-speed DC motor and an eccentric wheel. The motor is fixedly mounted on the outer wall of the water tank 2, and its output shaft is fixedly connected to the eccentric wheel. Multiple hinge holes are radially formed on the eccentric wheel. One end of a connecting rod is detachably hinged to one of these hinge holes, and the other end of the connecting rod is hinged to a oscillating pusher plate, which is vertically immersed in the water within the water tank 2. When the motor rotates, the eccentric wheel drives the connecting rod to reciprocate, causing the pusher plate to oscillate back and forth, thereby generating periodic waves within the water tank 2. The wave frequency can be changed by adjusting the motor speed, and the oscillation amplitude of the pusher plate can be changed by altering the hinge position of the connecting rod on the eccentric wheel, thus adjusting the wave height. In other feasible embodiments, the wave generation module 3 can also use a linear motor to directly drive the pusher plate, or a servo motor combined with a lead screw mechanism to achieve more precise wave control. In addition, two or more sets of pusher plates can be used to work together to simulate complex waveforms such as superimposed waves or standing waves.

[0024] The energy capture module 4 is used to capture wave energy and convert it into mechanical energy. This module includes a float made of lightweight closed-cell foam or hollow plastic, capable of floating on the water surface within the tank 2. The float is slidably fitted onto two vertical guide rods, which are vertically fixed to a bracket mounted above the tank 2, ensuring that the float moves only vertically and avoiding energy loss due to horizontal deviation. A crank-connecting rod mechanism is connected to the upper end of the float, converting the float's linear motion into rotational motion. In a preferred embodiment, the crank-connecting rod mechanism is made of stainless steel, and the crank radius and connecting rod length are designed to match the float's stroke and the generator speed. In other feasible embodiments, the energy capture module 4 can also use a rack and pinion mechanism instead of a crank-connecting rod, or a hydraulic cylinder to convert the float's motion into hydraulic energy to drive a hydraulic motor. Furthermore, the float's shape is not limited to cylindrical; it can also be spherical or conical to improve wave response sensitivity.

[0025] The power generation module 5 is connected to the energy harvesting module 4 to convert mechanical energy into electrical energy. In a preferred embodiment, the power generation module 5 is a miniature permanent magnet generator, with its rotor input shaft coaxially and fixedly connected to the output shaft of the crank-connecting rod mechanism. Energy storage capacitors are connected in parallel across the generator for voltage stabilization and energy storage. LED indicator lights are externally connected to the capacitors to visually display the power generation status. The rated power of the generator can be selected according to teaching needs; for example, 5 watts is sufficient to light up multiple LEDs and drive sensors and displays. In other feasible embodiments, the power generation module 5 can also use the relative motion of an electromagnetic induction coil and a magnet, or use a piezoelectric ceramic sheet to directly convert mechanical vibration into electrical energy, suitable for demonstrations of micro-wave energy harvesting.

[0026] The data display module 6 is electrically connected to the power generation module 5 and is used to collect and display power generation parameters. In a preferred embodiment, the data display module 6 includes a high-precision voltage sensor and a current sensor, which are connected in series to the generator's output circuit. The sensor's signal output terminal is connected to a liquid crystal digital display screen to display the output voltage and current values ​​in real time. The sensors can also collect power generation data and send the data to the edge computing intelligent control module 7 via a serial port. In other feasible embodiments, the data display module 6 can use an OLED screen or a touch screen and add a historical data curve plotting function, or synchronize data to a mobile app via a Bluetooth module.

[0027] The edge computing intelligent control module 7 is the control core of the entire device, electrically connected to the data display module 6, wave generation module 3, and energy capture module 4. This module incorporates an adaptive algorithm, enabling it to acquire real-time power generation data from the data display module 6 and dynamically optimize the wave generation parameters of the wave generation module 3 and the energy capture parameters of the energy capture module 4 based on this data. In a preferred embodiment, the edge computing intelligent control module 7 uses an STM32 series microcontroller or a Raspberry Pi Zero as the main control chip, running a lightweight deep Q-network reinforcement learning algorithm. This algorithm uses power generation as the reward signal and motor speed, pusher amplitude, and float depth as the action space, finding the optimal parameter combination for the current operating conditions through multiple rounds of iterative learning. Simultaneously, the module also integrates a wave spectrum fitting algorithm, capable of importing real ocean wave time-domain spectrum data to drive the wave generation module 3 to simulate complex operating conditions such as random waves, intermittent waves, or superimposed waves. In other feasible embodiments, the edge computing intelligent control module 7 can use an FPGA for faster real-time control, or adopt a cloud-edge collaborative architecture to offload some computing tasks to a cloud server. In addition, adaptive algorithms can also employ policy gradient reinforcement learning or evolutionary strategies to adapt to different hardware computing power constraints.

[0028] In summary, this embodiment constructs a complete wave energy power generation simulation and demonstration system through the organic combination of the support base 1, water tank 2, wave generation module 3, energy capture module 4, power generation module 5, data display module 6, and edge computing intelligent control module 7. The modules work together: wave generation module 3 generates controllable waves, energy capture module 4 converts wave energy into mechanical energy, power generation module 5 converts mechanical energy into electrical energy, data display module 6 provides real-time feedback of power generation parameters, and edge computing intelligent control module 7 intelligently adjusts the wave and capture parameters at the front end based on the feedback data, forming a closed-loop optimization. This system can not only simulate the complex wave environment of the real ocean but also autonomously learn and optimize power generation efficiency, significantly improving the simulation, scientific rigor, and intelligence level of teaching experiments, overcoming the shortcomings of traditional teaching aids with fixed working conditions and passive demonstrations.

[0029] In some embodiments, the adaptive algorithm is a lightweight deep Q-network reinforcement learning algorithm, configured to collect real-time power generation, wave frequency, and wave height data, and dynamically and adaptively adjust the motor speed and pusher amplitude of the wave generation module 3, as well as the float depth of the energy capture module 4, through iterative algorithm calculations. In this embodiment, the adaptive algorithm built into the edge computing intelligent control module 7 adopts a lightweight deep Q-network reinforcement learning algorithm. This algorithm uses power generation as an immediate reward signal and motor speed, pusher amplitude, and float depth as a continuous action space. It continuously updates the Q-value function through interaction with the environment, gradually approaching the optimal control strategy. In a preferred embodiment, the algorithm is deployed on an STM32F407 microcontroller, running a pruned and quantized neural network model. The input layer receives voltage and current sensor data from the data display module 6 and wave height values ​​read from the scale on the side wall of the water tank 2. The output layer provides the motor speed increment, target pusher amplitude, and target float depth for the next moment. Each iteration cycle is approximately 50 milliseconds, and it can converge to the optimal power generation efficiency point under the current wave conditions after dozens of trials. Another feasible implementation uses a Raspberry Pi Zero as the main controller, loads a pre-trained model using Python's TensorFlow Lite framework, and combines it with an online fine-tuning mechanism to continuously optimize parameters during experiments. A third implementation splits the algorithm into two parts: a fast decision network runs on the local MCU, while offline experience playback data is uploaded to a cloud server via WiFi for model updates, and the updated weights are then downloaded locally, balancing real-time performance and learning capabilities. In this way, the algorithm can collect real-time data on power generation, wave frequency, and wave height, and dynamically and adaptively adjust the motor speed and pusher amplitude of the wave generation module 3, as well as the float depth of the energy capture module 4, based on this data, achieving intelligent calibration of wave conditions and optimal matching of power generation efficiency. The beneficial effects of this embodiment are as follows: the lightweight deep Q-network reinforcement learning algorithm enables the demonstrator to have autonomous learning capabilities, automatically seeking optimization under various wave conditions without manual intervention, overcoming the shortcomings of traditional teaching aids that can only operate with fixed parameters; at the same time, the algorithm can simulate the random changes of wave parameters in the real ocean, making the experimental data closer to engineering practice and significantly improving the scientific nature and simulation of the teaching demonstration; in addition, since the algorithm model has been lightweighted, it can run smoothly on a low-cost embedded platform, reducing the overall cost and facilitating large-scale promotion and application.

[0030] In some embodiments, the edge computing intelligent control module 7 also incorporates a wave spectrum fitting algorithm, configured to simulate random waves, clutter, and / or superimposed waves in the real ocean. In this embodiment, the wave spectrum fitting algorithm built into the edge computing intelligent control module 7 is used to generate time-domain driving signals for real ocean waves. In a preferred embodiment, the algorithm is based on the classic JONSWAP spectrum model and pre-stores spectral parameters (such as significant wave height, peak frequency, and peak factor) for different sea states. During operation, the algorithm discretizes the target spectrum, assigns a random phase to each frequency component, and then synthesizes a continuous wave displacement time series through inverse fast Fourier transform. This series is updated every certain time interval (e.g., 10 seconds) as a modulation reference for the motor speed and pusher amplitude in the wave generation module 3, thereby generating random waves with realistic statistical characteristics in the water body in the tank 2. Another feasible implementation uses the Pierson-Moskowitz spectrum, which is suitable for fully developed wind and wave simulation. The algorithm decomposes the spectrum into multiple sinusoidal components and constructs irregular waves by superimposing sinusoidal waves with different amplitudes, frequencies, and random initial phases. The third implementation uses an autoregressive moving average (ARMA) model, training model parameters using measured ocean wave data, and then generating wave sequences consistent with the statistical characteristics of the training data in real time. This method has lower computational requirements for embedded platforms and is suitable for resource-constrained edge computing nodes. Through the above algorithm, the demonstrator can simulate common random waves, clutter, and superimposed wave conditions formed by the encounter of two or more waves in the real ocean. The beneficial effects of this embodiment are: the wave spectrum fitting algorithm fundamentally breaks through the limitation of traditional teaching aids that can only generate a single regular sine wave, making the experimental environment closer to actual ocean conditions. Students or researchers can observe wave behavior similar to real sea conditions in the laboratory, thereby deepening their understanding of the engineering challenges of wave energy power generation technology; at the same time, because the waves generated by the algorithm have controllable statistical characteristics, experimenters can systematically study the impact of different sea state levels on power generation efficiency and obtain more universal experimental conclusions; in addition, this algorithm works in conjunction with the aforementioned reinforcement learning adaptive algorithm. The former is responsible for providing a realistic disturbance environment, while the latter is responsible for finding the optimal control strategy in this environment. The combination of the two makes the entire device a complete intelligent simulation experimental platform capable of simulating the real ocean and autonomously optimizing.

[0031] In some embodiments, the edge computing intelligent control module 7 is also configured to support switching between manual mode and intelligent adaptive mode; In manual mode, users can manually adjust the motor speed of wave generation module 3, the hinge position of the connecting rod on the eccentric wheel, and the water depth of the float of energy capture module 4. In intelligent adaptive mode, the edge computing intelligent control module 7 autonomously adjusts the parameters of the wave generation module 3 and the energy capture module 4 based on the adaptive algorithm.

[0032] In this embodiment, the edge computing intelligent control module 7 is configured to support switching between manual mode and intelligent adaptive mode. The two modes are selected via a toggle switch or touch button on the support base 1. In a preferred embodiment, the main control chip of module 7 (such as STM32F407) runs two independent control threads: in manual mode, the chip shuts down the output channel of the adaptive algorithm and instead reads the parameter values ​​set by the user through the knob potentiometer and DIP switch, directly mapping these values ​​to the PWM duty cycle of the motor in wave generation module 3, the stepper motor drive signal corresponding to the hinge position of the eccentric wheel, and the servo angle of the float water depth adjustment mechanism in energy capture module 4. The user can finely adjust the motor speed through the knob, select different hinge hole positions on the eccentric wheel through the DIP switch, and set the positioning buckle position on the vertical slide of the float (executed by a micro electric actuator) through another knob, thereby realizing independent manual control of three sets of variables. Another feasible implementation is to draw a virtual dashboard on the LCD screen, and the user sets parameters by clicking or dragging the slider on the touch screen. Module 7 converts the touch coordinates into actual control quantities. The third implementation uses a Bluetooth connection to a mobile app, allowing users to adjust parameters on their phones. Module 7 receives and executes commands via a serial port. In intelligent adaptive mode, module 7 switches to the main loop of the adaptive algorithm, disabling the manual parameter setting channel. The algorithm automatically calculates the optimal motor speed, pusher blade amplitude, and float depth based on real-time power generation data and a built-in reinforcement learning model, and applies control through the same set of actuators (PWM, stepper motor, servo motor). When switching modes, module 7 saves a snapshot of the parameters in the current manual mode so that users can restore them at any time.

[0033] The beneficial effects of this embodiment are reflected in the following: the dual-mode design of manual and intelligent adaptive modes greatly broadens the applicability of the equipment. For primary and secondary school science popularization teaching or first-time users, the manual mode is intuitive and has a low barrier to entry. Students can directly adjust the knobs to intuitively feel the impact of motor speed, push plate swing, and float depth on power generation, which helps to establish a perceptual understanding of physical concepts. For university professional experiments or scientific research rehearsals, the intelligent adaptive mode demonstrates the application potential of artificial intelligence technology in the field of new energy. Students can observe how the algorithm learns autonomously and optimizes parameters, and understand the basic principles of reinforcement learning. The coexistence of the two modes allows the same equipment to complete both basic verification experiments and advanced exploratory experiments, avoiding the drawbacks of traditional teaching aids with limited functionality. In addition, the parameter snapshot saving mechanism during mode switching ensures the repeatability of experiments. Users can find a better operating point in manual mode and then switch to intelligent mode to observe whether the algorithm can further improve efficiency on this basis, thus intuitively comparing the differences between human experience and artificial intelligence. This design not only improves the flexibility of teaching but also provides convenient experimental conditions for algorithm performance evaluation.

[0034] In some embodiments, the energy harvesting module 4 further includes a bracket, a vertical guide slide, and a crank-connecting rod transmission mechanism; The support frame is installed above water tank 2; At least two vertical guide slide rods are fixed vertically to the bracket, and the float is slidably sleeved on the outside of the vertical guide slide rods and can float up and down along the vertical guide slide rods; The upper end of the float is connected to a crank-connecting rod transmission mechanism, which is configured to convert the linear reciprocating motion of the float into circular rotational motion and transmit it to the power generation module 5.

[0035] In this embodiment, the energy capture module 4 includes a bracket, vertical guide rods, and a crank-connecting rod transmission mechanism. The bracket, mounted above the water tank 2, is welded from aluminum alloy or stainless steel profiles and spans the edges of the bearing base 1 on both sides of the water tank 2, providing a stable mounting foundation for the entire energy capture module 4. At least two vertical guide rods are vertically fixed to the bracket. The surfaces of the guide rods are chrome-plated to reduce friction. A float is slidably fitted onto the outside of the vertical guide rod and can float up and down along the vertical guide rod. The float is made of closed-cell foam plastic or hollow polypropylene material, with a density less than water, enabling it to float stably on the water surface. The upper end of the float is connected to the crank-connecting rod transmission mechanism, which consists of a connecting rod and a crank. The lower end of the connecting rod is hinged to the top of the float, and the upper end of the connecting rod is hinged to one end of the crank. The other end of the crank is fixedly connected to the input shaft of the power generation module 5. The crank-connecting rod transmission mechanism is configured to convert the linear reciprocating motion of the float into circular rotational motion and transmit it to the power generation module 5.

[0036] In a preferred embodiment, two vertical guide rods are arranged parallel to each other on both sides of the float. The float engages with the guide rods via linear bearings to ensure smooth and unobstructed movement. The crank-connecting rod mechanism is made of 304 stainless steel, with a crank length of 30 mm and a connecting rod length of 80 mm, corresponding to a maximum float stroke of approximately 60 mm. The input shaft of the generator module 5 is connected to the crank's central shaft via a coupling to reduce concentricity errors. Another feasible implementation uses a single guide rod with an anti-rotation keyway. A keyway is provided on the inner side of the float to prevent rotation, resulting in a more compact structure. A third implementation replaces the crank-connecting rod mechanism with a rack and pinion mechanism. The rack is fixed to the side of the float, and the gear is connected to the shaft of the generator module 5. The linear motion is converted into rotational motion through gear meshing. This method has higher transmission efficiency but requires higher machining precision. Alternatively, a ball screw mechanism can be used. The float pushes the screw nut reciprocating, causing the screw to rotate and thus driving the generator, suitable for scenarios requiring higher torque.

[0037] The beneficial effects of this embodiment are as follows: the vertical guide rod design ensures that the float moves only in the vertical direction, eliminating energy loss and mechanical wear caused by horizontal deviation, and improving the stability and efficiency of energy capture. The crank-connecting rod transmission mechanism is simple in structure and low in cost, and can smoothly convert the low-speed, high-thrust float motion into high-speed rotational motion, adapting to the working characteristics of the micro permanent magnet generator. By adjusting the crank length or the position of the connecting rod fulcrum, the power generation needs under different wave amplitudes can be flexibly matched, enhancing the adaptability of the equipment. In addition, this mechanical transmission scheme is completely transparent and visible, allowing students to directly observe the causal relationship between the float's undulation and the generator's rotation, which helps to understand the conversion chain of wave energy-mechanical energy-electrical energy and enhances the intuitiveness of teaching. Compared with complex schemes using hydraulic or pneumatic transmission, the pure mechanical structure of this embodiment is easy to maintain, has a low failure rate, and is more suitable for frequently used teaching scenarios.

[0038] In some embodiments, the vertical guide slide is provided with several positioning buckles, configured to adjust the fixed position of the float on the vertical guide slide, thereby adjusting the water depth of the float. In this embodiment, the vertical guide slide is provided with several positioning buckles for adjusting the fixed position of the float on the vertical guide slide, thereby adjusting the water depth of the float. In a preferred embodiment, the positioning buckles adopt an elastic steel ball spring pin structure: an annular groove is machined every 5 mm along the axial direction of the vertical guide slide, and a spring-loaded steel ball is installed on the linear bearing sleeve inside the float. When the float moves to a certain groove position, the steel ball engages in the groove, producing a crisp click sound, achieving positioning and locking. The user only needs to push the float up or down to switch between adjacent grooves, without tools, making operation convenient. Another feasible implementation uses a threaded locking ring: a locking nut with a handle is installed on the top or bottom of the float. Loosening the nut allows the float to slide freely along the sliding rod, while tightening the nut fixes its position through friction. This method offers continuous stepless adjustment, but requires two hands for each adjustment. A third implementation uses magnetic positioning: multiple small magnets are embedded inside the sliding rod, and corresponding magnets are also installed inside the float sleeve. Magnetic attraction is used to position the float at a preset location. A slightly larger pushing force is required to overcome the magnetic force during adjustment, making it suitable for quick, coarse positioning. Regardless of the method, the positioning buckle allows the float's water depth to be adjusted in stages or continuously within a certain range (e.g., from 10 mm to 40 mm below the surface) to adapt to different wave intensities and power generation needs.

[0039] The beneficial effects of this embodiment are as follows: by adjusting the buoy's submersion depth, users can actively change the wave excitation force and damping characteristics experienced by the buoy, thereby directly affecting the energy capture efficiency. A shallower submersion depth is suitable for low-wave conditions, where the buoy responds sensitively but captures limited energy; a deeper submersion depth is suitable for high-wave conditions, where the buoy has a larger submerged volume and can absorb more wave energy, but may reduce the response speed due to excessive damping. This adjustable feature allows experimenters to systematically study the coupling relationship between the buoy's draft and power generation, deepening their understanding of the wave energy capture mechanism. Simultaneously, this design forms a closed loop with the adaptive algorithm in the edge computing intelligent control module 7: the algorithm can autonomously determine the optimal submersion depth based on real-time power generation data and achieve automated adjustment by unlocking and locking the positioning latch via a micro-electric push rod or servo motor, without manual intervention. This hardware-software integrated design not only enhances the flexibility of the experiment but also provides an additional control dimension for the reinforcement learning algorithm, helping to find the global optimum under a wider range of operating conditions. In addition, the positioning buckle has a simple and reliable mechanical structure that is not prone to failure even after long-term use, making it suitable for the frequent adjustment needs in a teaching environment.

[0040] In some embodiments, the adaptive algorithm is further configured to autonomously iteratively optimize the entry depth parameters of the buoy based on real-time power generation data. In this embodiment, the adaptive algorithm is further configured to autonomously iteratively optimize the entry depth parameters of the buoy based on real-time power generation data. In a preferred embodiment, this function is implemented using a lightweight deep Q-network reinforcement learning algorithm: the algorithm treats the entry depth of the buoy as an independent dimension in the continuous action space, alongside motor speed and pusher plate amplitude. During each iteration, the algorithm outputs an action command containing an adjustment amount for the entry depth based on a state vector composed of the current power generation, wave frequency, and wave height data. This command drives a micro electric actuator or stepper motor through the edge computing intelligent control module 7, moving the buoy along the vertical guide slide to a new depth position. After multiple attempts, the algorithm gradually learns the optimal entry depth that maximizes power generation under different wave conditions. Another feasible implementation uses a Bayesian optimization method: the algorithm maintains a Gaussian process surrogate model, updates the model based on the observed power generation after each experiment, and selects the next water depth to be tested using the acquisition function (if desired to be improved). This method is superior to reinforcement learning in terms of sample efficiency and is suitable for scenarios with a limited number of experiments. A third implementation uses a simple hill-climbing algorithm: the algorithm explores several candidate depths in a small neighborhood near the current water depth, selects the one with the highest power generation as the new position, and repeats this process until convergence. This method has the lowest computational cost and is suitable for embedded platforms with extremely low computing power.

[0041] The beneficial effects of this embodiment are reflected in the fact that by incorporating the buoy's submersion depth into the optimization scope of the adaptive algorithm, the energy capture module 4 possesses full-dimensional autonomous adjustment capabilities. In traditional teaching, the buoy's submersion depth is usually set by teachers based on experience, making it difficult to guarantee that it is optimal under different wave conditions. After the algorithm autonomously iterates and optimizes, the equipment can automatically find the most suitable draft depth for the current wave conditions without human intervention, significantly improving power generation efficiency. At the same time, this process itself is a vivid demonstration of reinforcement learning algorithms: students can observe how the algorithm gradually approaches the optimal value through trial and error, intuitively understanding the decision-making logic of artificial intelligence. In addition, since the submersion depth directly affects the coupling degree between the buoy and the waves, optimizing this parameter can also reduce energy loss caused by excessive submersion or insufficient capture caused by undersubmersion, extend the service life of the equipment, and improve the repeatability of experimental data. This optimization approach from local to global also provides an excellent case study for students to understand multi-parameter collaborative optimization.

[0042] In some embodiments, the wave generation module 3 includes a speed-regulating drive motor, an eccentric wheel, a connecting rod, and a swinging pusher plate; The speed-regulating drive motor is fixedly installed on the outer wall of the water tank 2, and the output shaft of the speed-regulating drive motor is fixedly connected to the eccentric wheel. The eccentric wheel is provided with several radially distributed hinge holes. One end of the connecting rod is detachably hinged to one of the hinge holes, and the other end of the connecting rod is hinged to the swing push plate. The oscillating push plate is vertically installed inside the water tank 2; The wave height can be adjusted by selecting different hinge holes and adjusting the swing amplitude of the oscillating pusher plate.

[0043] In this embodiment, the wave generation module 3 adopts a combined structure of a speed-regulating drive motor, an eccentric wheel, a connecting rod, and a oscillating pusher plate. The speed-regulating drive motor is fixedly installed on the outer wall of the water tank 2, and the output shaft of the motor is fixedly connected to the central hole of the eccentric wheel. Multiple hinge holes are radially formed on the eccentric wheel, and the distribution radius of these holes varies. One end of the connecting rod is detachably hinged to one of the hinge holes via a pin, and the other end of the connecting rod is hinged to the upper or middle part of the oscillating pusher plate. The oscillating pusher plate is vertically installed inside the water tank 2, with its lower end immersed in the water. By selecting hinge holes at different radial positions on the eccentric wheel, the swing stroke of the connecting rod can be changed, thereby adjusting the swing amplitude of the oscillating pusher plate and ultimately controlling the wave height.

[0044] In a preferred embodiment, the speed-regulating drive motor is a 12V DC geared motor with a rated speed of 0-300 rpm, which, in conjunction with a synchronous pulley and belt, achieves stepless speed regulation. The eccentric wheel is machined from No. 45 steel, with a diameter of 80 mm and a thickness of 10 mm, and has four hinge holes evenly distributed radially, at distances of 15 mm, 25 mm, 35 mm, and 45 mm from the center, respectively. The connecting rod is made of hollow aluminum alloy tubing, 120 mm long, with spherical bearings installed at both ends to reduce friction. The oscillating pusher plate is made of acrylic sheet, with a width consistent with the inner width of the water tank 2 (e.g., 240 mm), a height of 260 mm, and a bottom distance of 10 mm from the bottom surface of the water tank 2. When the connecting rod is connected to the hinge hole closest to the center (15 mm), the pusher plate oscillates with an amplitude of approximately ±5 degrees, generating a wave height of approximately 5 mm; when connected to the hinge hole furthest from the center (45 mm), the oscillation amplitude is approximately ±20 degrees, and the wave height can reach 30 mm. Another feasible implementation uses a stepper motor instead of a DC motor, achieving more precise speed regulation through driver microstepping control. Simultaneously, up to six hinge holes are added to the eccentric wheel to provide finer swing amplitude levels. A third implementation replaces the eccentric wheel with a cam mechanism. The cam profile curve can be customized according to the desired wave waveform. The connecting rod connects to the cam follower, enabling the generation of complex, non-sinusoidal oscillation trajectories, thereby simulating more realistic ocean wave patterns.

[0045] The beneficial effects of this embodiment are as follows: The design of multiple radial hinge holes on the eccentric wheel enables mechanical, graded adjustment of wave height, allowing for rapid switching between various wave height conditions without replacing parts. This is simple to operate and cost-effective. The combination of the speed-regulating drive motor and the eccentric wheel allows for independent control of wave frequency (by adjusting the motor speed) and wave height (by selecting the hinge holes), with no interference between them. This provides two independent variables for experiments, facilitating students' investigation of the individual effects of frequency and wave height on power generation efficiency. Simultaneously, this mechanical structure is completely transparent and visible, allowing students to intuitively observe how the rotation of the eccentric wheel translates into the oscillation of the pusher plate, and then into waves on the water surface, aiding in understanding the fundamental principles of mechanical transmission and fluid dynamics. Furthermore, the structure is robust and durable, resistant to damage even with frequent use, making it suitable for high-intensity operation in teaching environments. When used in conjunction with the edge computing intelligent control module 7, the hinge position of the connecting rod can be automatically switched via a micro servo motor, achieving automatic wave height adjustment and further enhancing the equipment's intelligence.

[0046] In some embodiments, the data display module 6 includes a voltage sensor, a current sensor, and a liquid crystal digital display screen; The voltage sensor and the current sensor are connected in series to the output circuit of the power generation module 5. The signal output terminals of the voltage sensor and the current sensor are both connected to the LCD digital display screen and are configured to display the output voltage and current values ​​in real time. The voltage and current sensors are also connected to the edge computing intelligent control module 7 and are configured to provide real-time power generation data to the edge computing intelligent control module 7.

[0047] In this embodiment, the data display module 6 includes a voltage sensor, a current sensor, and a liquid crystal display screen. The voltage sensor and the current sensor are connected in series to the output circuit of the power generation module 5. The signal output terminals of both the voltage sensor and the current sensor are connected to the liquid crystal display screen to display the output voltage and current values ​​in real time. Simultaneously, the voltage sensor and the current sensor are also connected to the edge computing intelligent control module 7 to provide it with real-time power generation data.

[0048] In a preferred embodiment, the voltage sensor uses an ACS712 Hall effect current sensor with a range of ±5 amps, and the output voltage signal is proportional to the current. The current sensor uses a resistor divider voltage detection circuit, which, after conditioning with a precision operational amplifier, outputs a 0-3.3V analog signal. The outputs of the two sensors are connected to the analog input pins of a 128×64 pixel monochrome graphic dot matrix LCD with a built-in character library, capable of simultaneously displaying voltage, current, and instantaneous power values, with a refresh rate of 10 times per second. The analog signals from the sensors are simultaneously sent to the analog-to-digital conversion channel of the edge computing intelligent control module 7, with a sampling accuracy of 12 bits and a sampling rate of 100 Hz, for the adaptive algorithm to calculate power generation and efficiency indicators. Another feasible implementation uses an INA219 digital power monitoring chip, which directly outputs voltage, current, and power values ​​via an I2C interface, eliminating the need for analog conditioning circuitry. Both the LCD and the edge computing module read data via the I2C bus, simplifying wiring. The third implementation uses a high-precision multimeter module to convert sensor signals into RS232 serial data. The LCD screen and edge computing module receive the data via the serial port. This method has strong anti-interference capabilities and is suitable for long-distance transmission. Furthermore, the LCD digital display can be replaced with a TFT color touchscreen, which can not only display numerical values ​​but also plot voltage and current curves over time, enhancing the visualization effect.

[0049] The beneficial effects of this embodiment are as follows: voltage and current sensors collect electrical parameters of the power generation module 5 in real time and present them intuitively on an LCD digital display screen. Students can observe the impact of wave intensity changes on power generation in real time, deepening their understanding of the energy conversion process. Simultaneously, this sensor data is directly fed into the edge computing intelligent control module 7, forming a closed-loop feedback input for the adaptive algorithm. This allows the algorithm to dynamically adjust wave-generating and energy-capturing parameters based on actual power generation performance, achieving optimal efficiency matching. The presence of the data display module 6 not only meets the intuitive requirements of teaching demonstrations but also provides the necessary data foundation for intelligent control. The two complement each other, enhancing the overall intelligence level of the equipment and the scientific rigor of the experiment. Furthermore, the alternative solutions for multiple sensor interface methods ensure system compatibility with different hardware platforms, reducing manufacturing costs and maintenance difficulty.

[0050] In some embodiments, the support base 1 is a hollow box structure, and the support base 1 houses electrical connection lines and edge computing intelligent control module 7. A main power switch is installed on the outside of the support base 1, which is configured to control the overall power-on and power-off of the wave generation module 3, the data display module 6 and the edge computing intelligent control module 7.

[0051] The support base 1 adopts a hollow box structure, internally housing electrical connection lines and the edge computing intelligent control module 7. In a preferred embodiment, the support base 1 is injection molded from ABS engineering plastic, with external dimensions of 450 mm in length, 300 mm in width, and 80 mm in height, and a wall thickness of 3 mm. The interior of the base is divided into three independent areas by cross-shaped plastic partitions: the left area holds the main control circuit board of the edge computing intelligent control module 7, the middle area has wiring and terminal blocks installed, and the right area houses the power transformer and relays. Wiring holes are provided in the partitions, and the wires are protected with rubber coil guards. A silicone sealing gasket is affixed to the top surface of the base where it contacts the water tank 2 to prevent water from seeping into the interior. The main power switch is a rocker switch with indicator lights, installed on the right side of the front of the support base 1, with a rated current of 10 amps, controlling the on / off state of the 220V AC input, thereby controlling the switching power supply module to provide unified power to the speed-regulating drive motor of the wave generation module 3, the LCD screen and sensors of the data display module 6, and the edge computing intelligent control module 7. Another feasible implementation uses sheet metal bending technology. The base is made of 1.2 mm thick cold-rolled steel plate with electrostatic powder coating. An internal DIN rail is installed, and the controller of the edge computing intelligent control module 7 is also mounted on the rail. The wiring terminals also use rail-mounted terminal blocks for easy disassembly and maintenance. The main power switch is replaced with a key switch with a lock; only by inserting the key can the machine be turned on, preventing accidental operation by students. A third implementation uses a 3D-printed resin shell with a completely customized internal structure. The main control board, sensor interface board, and power module are stacked using copper pillars, making full use of vertical space to reduce volume. The main power switch is a touch switch with LED backlighting, resulting in a more modern appearance.

[0052] The beneficial effects of this embodiment are as follows: The hollow box structure centrally houses all electrical connection lines and the edge computing intelligent control module 7 inside the support base 1, avoiding the safety hazards caused by exposed wires and preventing short circuits caused by accidental water splashes during experiments. Simultaneously, this integrated design makes the entire device neat in appearance, occupies little desktop space, and is easy to move and store, making it ideal for shared environments such as classrooms and laboratories. The main power switch enables one-button start / stop control of the entire system, eliminating the need for users to separately turn off the independent power supplies of the motor, sensors, and display screen, simplifying the operation process. This is particularly suitable for the frequent power on / off needs of teachers during rapid demonstrations or students conducting group experiments in primary and secondary school classrooms. Furthermore, the internal partition partitions and cable routing holes ensure neat and orderly wiring. During maintenance, simply opening the base plate allows for clear identification of the connections between modules, reducing the difficulty of later maintenance. The availability of various shell materials and switch types allows manufacturers to flexibly choose according to cost and positioning, ensuring both functional integrity and cost-effectiveness and aesthetics.

[0053] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0054] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0055] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A wave energy generation simulator equipped with an adaptive algorithm, characterized in that, include: Support base (1); Water tank (2) is fixedly installed on the bearing base (1); Wave generation module (3), the drive part of the wave generation module (3) is disposed outside the water tank (2), the execution part of the wave generation module (3) extends into the water tank (2), and the wave generation module (3) is configured to generate waves in the water body in the water tank (2); The energy capture module (4) includes a float that can float on the surface of the water in the tank (2), and the energy capture module (4) is configured to capture wave energy and convert the wave energy into mechanical energy; The power generation module (5), connected to the energy capture module (4), is configured to convert the mechanical energy into electrical energy; The data display module (6) is electrically connected to the power generation module (5) and is configured to collect and display power generation parameters; The edge computing intelligent control module (7) is electrically connected to the data display module (6), the wave generation module (3), and the energy capture module (4), respectively; The edge computing intelligent control module (7) has a built-in adaptive algorithm and is configured to obtain real-time power generation data from the data display module (6) and dynamically optimize the wave generation parameters of the wave generation module (3) and the energy capture parameters of the energy capture module (4) based on the real-time power generation data, so as to achieve intelligent calibration of wave conditions and optimal matching of power generation efficiency.

2. The demonstrator according to claim 1, characterized in that, The adaptive algorithm is a lightweight deep Q-network reinforcement learning algorithm, which is configured to collect power generation, wave frequency and wave height data in real time, and dynamically and adaptively adjust the motor speed and push plate swing of the wave generation module (3) and the water depth of the float of the energy capture module (4) through algorithm iteration.

3. The demonstrator according to claim 1, characterized in that, The edge computing intelligent control module (7) also has a built-in wave spectrum fitting algorithm, which is configured to simulate random waves, clutter and / or superimposed waves in the real ocean.

4. The demonstrator according to claim 1, characterized in that, The edge computing intelligent control module (7) is also configured to support switching between manual mode and intelligent adaptive mode; In the manual mode, the user is allowed to manually adjust the motor speed of the wave generation module (3), the hinge position of the connecting rod on the eccentric wheel, and the water depth of the float of the energy capture module (4); In the intelligent adaptive mode, the edge computing intelligent control module (7) autonomously controls the parameters of the wave generation module (3) and the energy capture module (4) according to the adaptive algorithm.

5. The demonstrator according to claim 1, characterized in that, The energy harvesting module (4) also includes a bracket, a vertical guide slide, and a crank-connecting rod transmission mechanism; The bracket is mounted above the water tank (2); At least two of the vertical guide slide rods are vertically fixed to the bracket, and the float is slidably sleeved on the outside of the vertical guide slide rods and can float up and down along the vertical guide slide rods; The upper end of the float is connected to the crank-connecting rod transmission mechanism, which is configured to convert the linear reciprocating motion of the float into circular rotational motion and transmit it to the power generation module (5).

6. The demonstrator according to claim 5, characterized in that, The vertical guide slide is provided with several positioning buckles, which are configured to adjust the fixed position of the float on the vertical guide slide, thereby adjusting the water depth of the float.

7. The demonstrator according to claim 5, characterized in that, The adaptive algorithm is also configured to autonomously iteratively optimize the water depth parameters of the buoy based on real-time power generation data.

8. The demonstrator according to claim 1, characterized in that, The wave generation module (3) includes a speed-regulating drive motor, an eccentric wheel, a connecting rod, and a swing pusher plate; The speed-regulating drive motor is fixedly installed on the outer wall of the water tank (2), and the output shaft of the speed-regulating drive motor is fixedly connected to the eccentric wheel; The eccentric wheel is provided with a plurality of radially distributed hinge holes, one end of the connecting rod is detachably hinged to one of the hinge holes, and the other end of the connecting rod is hinged to the swing push plate; The swing pusher plate is vertically installed inside the water tank (2); The wave height can be adjusted by selecting different hinge holes and adjusting the swing amplitude of the swinging pusher plate.

9. The demonstrator according to claim 1, characterized in that, The data display module (6) includes a voltage sensor, a current sensor, and a liquid crystal digital display screen; The voltage sensor and the current sensor are connected in series to the output circuit of the power generation module (5). The signal output terminals of the voltage sensor and the current sensor are both connected to the liquid crystal digital display screen and are configured to display the output voltage and current values ​​in real time. The voltage sensor and the current sensor are also connected to the edge computing intelligent control module (7) and are configured to provide the real-time power generation data to the edge computing intelligent control module (7).

10. The demonstrator according to claim 1, characterized in that, The bearing base (1) is a hollow box structure, and the bearing base (1) houses the electrical connection lines and the edge computing intelligent control module (7); The main power switch is provided on the outside of the bearing base (1), which is configured to control the overall power-on and power-off of the wave generation module (3), the data display module (6) and the edge computing intelligent control module (7).

Images