Charging Control Method, Charging System, Smart Lock and Storage Medium

Abstract

The present invention relates to the field of charging control technology, specifically a charging control method, a charging system, a smart lock, and a storage medium. The steps of the charging control method include: constructing a multidimensional state space of the battery charging system of the charging system; designing a charging path planning model based on the ant colony optimization algorithm, and treating the charging process as a process of ants searching for the optimal path in the multidimensional state space; collecting the working status information and environmental parameters of the charging system in real time; converting the optimal charging path into a specific charging control strategy, executing the charging control strategy in real time, and continuously monitoring the charging effect. The charging control method proposed in the present invention effectively solves the adaptive optimization problem of the smart lock battery charging system by constructing a multidimensional state space and applying the ant colony optimization algorithm.

Description

Technical Field

[0001] The present invention relates to the field of charging control technology, and specifically to a charging control method, a charging system, a smart lock, and a storage medium. Background Art

[0002] Charging management of batteries or charging systems has always been a research topic in the industry. Traditional charging methods mostly use fixed charging strategies that cannot be adjusted according to actual usage and environmental conditions. As a result, charging efficiency cannot be changed according to demand, affecting battery life.

[0003] Many power-consuming devices, such as smart locks, are subject to complex and variable operating environments. Factors such as temperature fluctuations, usage frequency fluctuations, and network connectivity can affect the charging process. Furthermore, different user habits can lead to varying energy demands. These factors make it difficult to design a universal and efficient charging strategy.

[0004] In existing technologies, optimized charging strategies usually only consider a single or a few factors, which cannot fully capture the complexity of smart lock charging systems. Summary of the Invention

[0005] The present invention provides a charging control method, a charging system, a smart lock and a storage medium to solve the technical problem of low charging efficiency in existing charging methods due to the influence of adaptive adjustment capability.

[0006] The technical solution of the present invention to solve the above technical problems is as follows:

[0007] A charging control method is provided, which is applied to a charging system. The charging control method includes the following steps:

[0008] Constructing a multidimensional state space of a battery charging system for the charging system;

[0009] A charging path planning model is designed based on the ant colony optimization algorithm, which regards the charging process as a process in which ants search for the optimal path in a multi-dimensional state space.

[0010] Real-time collection of charging system working status information and environmental parameters;

[0011] The collected information is fed into a pre-trained deep neural network to predict the usage patterns and energy requirements of the charging system.

[0012] Based on the prediction results, the parameters in the ant colony optimization algorithm are dynamically adjusted, and the ant colony optimization algorithm is executed in the adjusted multi-dimensional state space to find the optimal charging path;

[0013] Convert the optimal charging path into a specific charging control strategy, execute the charging control strategy in real time, and continuously monitor the charging effect.

[0014] In another aspect, a charging system is provided, the charging system comprising:

[0015] Multi-dimensional state space construction module, used to construct the multi-dimensional state space of the smart lock battery charging system;

[0016] The ant colony optimization path planning module is used to design a charging path planning model based on the ant colony optimization algorithm, which regards the charging process as a process in which ants search for the optimal path in a multi-dimensional state space;

[0017] Real-time data acquisition module, used to collect the working status information and environmental parameters of the smart lock in real time;

[0018] A deep neural network prediction module that receives the collected information and predicts the usage pattern and energy demand of the smart lock;

[0019] Parameter dynamic adjustment module, used to dynamically adjust the parameters in the ant colony optimization algorithm based on the prediction results;

[0020] An optimization algorithm execution module is used to execute an ant colony optimization algorithm in the adjusted multi-dimensional state space to find the optimal charging path;

[0021] A charging strategy conversion module is used to convert the optimal charging path into a specific charging control strategy;

[0022] Charging control execution module, used to execute charging control strategy in real time;

[0023] Closed-loop feedback monitoring module, used to continuously monitor charging results and provide feedback.

[0024] In another aspect, a smart lock is provided, comprising the above charging system, the smart lock further comprising:

[0025] Integrated micro-supercapacitor for peak load management and fast response;

[0026] Flexible solar cells and piezoelectric materials for ambient energy harvesting;

[0027] Graphene-based heat dissipation structure for improving thermal management efficiency;

[0028] Programmable charging interface to support multiple charging protocols and bidirectional charging function.

[0029] On the other hand, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above charging control method are implemented.

[0030] The beneficial effects of the present invention are:

[0031] The charging control method proposed in this paper effectively solves the adaptive optimization problem of smart lock battery charging systems by constructing a multidimensional state space and applying the ant colony optimization algorithm. This method can capture and analyze multiple influencing factors in real time, including battery status, environmental conditions, and usage patterns, to dynamically adjust the charging strategy.

[0032] By abstracting the charging process into the problem of ants finding the optimal path in multi-dimensional space, it can not only handle high-dimensional state space, but also quickly converge to a high-quality solution with limited computing resources.

[0033] Combined with the predictive capabilities of deep neural networks, the method of this invention can accurately predict the usage patterns and energy requirements of smart locks, further improving the adaptive charging strategy. By dynamically adjusting algorithm parameters and executing the optimized charging strategy in real time, this method continuously optimizes the charging process, effectively improving charging efficiency, extending battery life, and adapting to a variety of complex usage scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of a charging control method according to an embodiment of the present invention;

[0035] Figure 2 Schematic diagram of ant colony optimization path planning in one embodiment of the present invention Figure 1 ;

[0036] Figure 3 Schematic diagram of ant colony optimization path planning in one embodiment of the present invention Figure 2 ;

[0037] Figure 4 A diagram showing the steps of charging path planning in one embodiment of the present invention;

[0038] Figure 5 This is a diagram of the architecture of a charging system according to an embodiment of the present invention;

[0039] In the accompanying drawings, the components represented by the reference numerals are as follows:

[0040] 100. Multi-dimensional state space construction module; 200. Ant colony optimization path planning module; 300. Real-time data acquisition module; 400. Deep neural network prediction module; 500. Parameter dynamic adjustment module; 600. Optimization algorithm execution module; 700. Charging strategy conversion module; 800. Charging control execution module; 900. Closed-loop feedback monitoring module. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0042] The term "comprise" and any variations thereof in the specification and claims of this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to the process, method, product, or apparatus. In addition, the use of "and / or" in the specification and claims to indicate at least one of the connected objects, such as A and / or B, means that A alone, B alone, and both A and B are included.

[0043] In the embodiments of the present invention, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present invention should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0044] The present invention provides the following preferred embodiments:

[0045] Example 1

[0046] In order to solve the problem of adaptive optimization of the charging system in a complex and changing environment, this embodiment proposes various steps of a charging control method to achieve efficient and intelligent charging management.

[0047] like Figure 1 As shown, the steps of the charging control method include:

[0048] S101. Construct a multi-dimensional state space of a battery charging system.

[0049] S102. Design a charging path planning model based on the ant colony optimization algorithm, and regard the charging process as a process in which ants search for the optimal path in a multi-dimensional state space.

[0050] S103 : Collecting working status information and environmental parameters of the charging system in real time.

[0051] S104: Input the collected information into a pre-trained deep neural network to predict the usage pattern and energy demand of the charging system.

[0052] S105 , dynamically adjusting parameters in the ant colony optimization algorithm based on the prediction results, and executing the ant colony optimization algorithm in the adjusted multi-dimensional state space to find the optimal charging path.

[0053] S106: Convert the optimal charging path into a specific charging control strategy, execute the charging control strategy in real time, and continuously monitor the charging effect.

[0054] Specifically, this embodiment first constructs a multidimensional state space for the charging system. This multidimensional state space encompasses multiple dimensions, including battery charge level, charging current, charging voltage, ambient temperature, usage frequency, unlock times, network connection status, data transmission frequency, electromagnetic interference intensity, and battery health status. By constructing such a comprehensive state space, the operating status of the charging system can be more accurately described, providing a foundation for subsequent optimization.

[0055] Furthermore, this embodiment designs a charging path planning model based on the ant colony optimization algorithm. This model abstracts the charging process into a process in which ants search for the optimal path in a multi-dimensional state space. The characteristic of the ant colony optimization algorithm is that it can quickly find an approximate optimal solution in a complex solution space, and is suitable for handling multi-variable and multi-objective optimization problems of charging systems. Figure 2 As shown in the figure, a 2D space is created, which contains randomly distributed "states", which can be understood as different charging parameter combinations, defining the starting point (charging start) and the end point (charging end), and using the ant colony optimization algorithm to find the optimal path between these states. Figure 2 (a) to (d) show the results of 1, 5, 20, and 50 iterations, respectively. The dots represent different charging states, the dotted line connects the starting point (charging start) and the end point (charging end), the line segments represent the connection between states, and the transparency of the line represents the pheromone intensity. As the number of iterations increases, certain paths will become more obvious, indicating that these paths are more optimal charging strategies. Figure 3 As shown, a 3D space is used to represent the multidimensional state space, with the three dimensions of battery charge, charging current, and temperature arbitrarily chosen. The X-axis represents the battery charge (%), the Y-axis represents the charging current (A), and the Z-axis represents the temperature (°C). The dots represent different charging states, with the two larger dots representing the starting point (start of charging) and the end point (end of charging). The transparency of the dots represents the pheromone intensity. As the number of iterations increases, some dots become opaque, indicating that these dots represent the more optimal charging strategy.

[0056] Furthermore, this embodiment employs a real-time data acquisition mechanism. Through a high-precision sensor network, the system can acquire real-time information about the charging system's operating status and environmental parameters. This data includes, but is not limited to, real-time battery charge, temperatures of key components, unlock counts, network connection status, data transmission frequency, and ambient temperature, humidity, and electromagnetic interference intensity.

[0057] As you can understand, the collected information is fed into a pre-trained deep neural network. This neural network, trained on extensive historical data, can accurately predict usage patterns and future energy demands for smart locks that incorporate charging systems. This predictive capability allows the charging system to proactively prepare for potential malfunctions caused by insufficient power.

[0058] Furthermore, based on the prediction results, the system dynamically adjusts parameters within the ant colony optimization algorithm. These parameters include the pheromone volatility rate and the balance factor between global and local searches. By adjusting these parameters, the ant colony optimization algorithm can better adapt to current charging needs and environmental conditions. In this adjusted multidimensional state space, the system executes the ant colony optimization algorithm to find the optimal charging path.

[0059] In this embodiment, the optimal charging path must balance multiple objectives, including minimizing charging time, maximizing battery life, ensuring sufficient charge during predicted peak usage periods, and controlling temperature rise during charging. The system converts the found optimal path into a specific charging control strategy, which includes designing a pulse charging waveform, determining multi-stage charging current and voltage curves, and formulating a temperature control strategy.

[0060] Furthermore, this embodiment employs a closed-loop feedback mechanism to continuously monitor charging performance. The system uses high-speed data acquisition equipment to monitor minute changes in charging current, voltage, and temperature in real time. It uses Fourier transform analysis to analyze the harmonic components of the charging current to assess charging efficiency, and employs a Kalman filter algorithm to estimate and predict the charging status in real time.

[0061] The benefit of this embodiment lies in the adaptive optimization of the charging system achieved by constructing a multidimensional state space and applying the ant colony optimization algorithm. This method can capture and analyze multiple influencing factors in real time, dynamically adjusting the charging strategy, effectively improving charging efficiency, extending battery life, and adapting to various complex usage scenarios. Furthermore, through the prediction capabilities of deep neural networks and a closed-loop feedback mechanism, this method achieves continuous optimization of the charging process.

[0062] Example 2

[0063] To address the multi-dimensional information processing and optimization challenges of smart lock charging systems in complex environments, this embodiment further refines the method for constructing a multi-dimensional state space. By precisely defining and quantifying the key factors that influence the charging process, this embodiment aims to provide a comprehensive and highly adaptable charging control framework.

[0064] The multi-dimensional state space in this embodiment includes, but is not limited to, the following dimensions: battery level, charging current, charging voltage, ambient temperature, usage frequency, number of unlocks, network connection status, data transmission frequency, electromagnetic interference intensity, and battery health status. Each dimension represents a factor that affects the charging process.

[0065] Furthermore, the battery charge dimension is monitored in real time using a high-precision coulomb counter with an accuracy of up to 0.1%. The charging current and charging voltage dimensions are achieved through a high-speed sampling ADC with a sampling rate of up to 1MHz to capture transient changes. The ambient temperature dimension is measured using a PT100 platinum resistance temperature sensor with a temperature range of -50°C to 150°C and an accuracy of ±0.1°C.

[0066] It's important to understand that the usage frequency and unlock count dimensions reflect the actual usage of the smart lock. These two dimensions are implemented using embedded counters, synchronized with the real-time clock module to provide timestamp information. The network connection status dimension is quantified by monitoring the signal strength and connection stability of the wireless module, using the received signal strength indicator (RSSI) value as the primary indicator.

[0067] It's understandable that the data transmission frequency dimension reflects the intensity of interaction between the smart lock and the external system. This dimension is quantified by counting the number of data packets transmitted per unit time and smoothed using a sliding window algorithm to eliminate the impact of short-term fluctuations.

[0068] Furthermore, the electromagnetic interference intensity dimension is measured using a broadband EMI receiver with a frequency range of 9kHz to 1GHz. The introduction of this dimension helps to assess the electromagnetic compatibility of the charging environment and thus optimize the charging strategy to reduce potential interference effects.

[0069] Furthermore, the battery health dimension is a comprehensive indicator calculated from multiple parameters. These parameters include battery internal resistance, capacity decay rate, and charge and discharge efficiency. This embodiment uses electrochemical impedance spectroscopy (EIS) technology to regularly measure the battery's internal resistance and, combined with historical charge and discharge data, uses a machine learning algorithm to predict the battery's remaining life.

[0070] It's important to note that these dimensions don't exist independently, but rather interact in complex ways. For example, ambient temperature affects battery charging efficiency and health; usage frequency and unlocking times affect battery drain; and network connectivity and data transmission frequency influence the power consumption characteristics of smart locks.

[0071] Furthermore, this embodiment integrates this multi-dimensional information by introducing a data fusion algorithm. Specifically, a Kalman filter is used to suppress noise and estimate the state of each dimension of data. Principal component analysis (PCA) is then used to reduce the data dimensionality and extract the most representative features. Finally, a deep neural network model is used to analyze and predict the processed data, providing a basis for developing charging control strategies.

[0072] Furthermore, this embodiment can also introduce an adaptive sampling mechanism. For dimensions that change rapidly (such as charging current and charging voltage), a higher sampling frequency is used; for dimensions that change slowly (such as battery health status), a lower sampling frequency is used to balance system resource consumption and data accuracy.

[0073] The benefit of this embodiment is that, by constructing such a multidimensional state space, the charging control system can comprehensively perceive and analyze the operating status of the smart lock and its environment. This multidimensional information acquisition and processing method provides an accurate data foundation for subsequent charging path planning and optimization, helping to develop more precise and efficient charging strategies.

[0074] Example 3

[0075] In order to solve the problem of multi-objective optimization in charging path planning, this embodiment further refines the charging path planning model based on the ant colony optimization algorithm, and realizes a more flexible and efficient charging control method by introducing a multi-dimensional evaluation function and a dynamic weight adjustment mechanism.

[0076] like Figure 4 As shown in FIG, the steps of designing a charging path planning model based on the ant colony optimization algorithm include:

[0077] S301. Define a path evaluation function F(path). The evaluation function F(path) includes the following evaluation parameters: charging time T, battery health status change ΔH, charging energy efficiency E, temperature change ΔTemp, and weight coefficients w1, w2, w3, and w4.

[0078] S302 : For each possible charging path, calculate the charging time T, evaluate the battery health status change ΔH, calculate the charging energy efficiency E, and measure the temperature change ΔTemp.

[0079] S303, through the formula Calculate the evaluation score.

[0080] S304: Dynamically adjust the weight coefficient according to different charging scenarios and priorities.

[0081] S305: Select the path with the highest evaluation score as the optimal charging path.

[0082] In this embodiment, the charging path planning model defines a comprehensive path evaluation function F(path). This evaluation function comprehensively considers multiple key parameters during the charging process, including charging time T, change in battery health ΔH, charging energy efficiency E, and temperature change ΔTemp. These parameters correspond to considerations of efficiency, safety, energy utilization, and thermal management during the charging process.

[0083] Furthermore, to adapt the evaluation function to different charging scenarios and priority requirements, weight coefficients w1, w2, w3, and w4 are introduced. These weight coefficients can be dynamically adjusted based on specific application requirements, enabling flexible configuration of charging strategies. For example, in a fast-charging scenario, the weight w1 corresponding to the charging time T can be increased; while in a scenario where battery life is a priority, the weight w2 corresponding to the change in battery health status ΔH can be increased.

[0084] Furthermore, in practical applications, for each possible charging path, the system performs the following steps: First, the time T required to complete charging is calculated. This can be achieved by simulating the charging process or using a predictive model based on historical data. Second, the change in battery health, ΔH, is evaluated. This typically involves analyzing the battery's cycle life model and degradation mechanisms. Then, the charging energy efficiency, E, is calculated, which requires considering the energy loss and conversion efficiency during the charging process. Finally, the temperature change, ΔTemp, during charging is measured or predicted, which is crucial for preventing battery overheating and optimizing the thermal management system.

[0085] It is important to understand that the evaluation score is calculated using a comprehensive weighted approach to convert parameters of different dimensions to the same scale. This normalization process ensures comparability between different parameters, allowing the evaluation score to truly reflect the overall performance of the charging path.

[0086] It's understood that dynamic adjustment of weight coefficients can be achieved through pre-set scenario modes or automatically optimized through machine learning algorithms based on historical charging data and user feedback. For example, the system can identify whether the current power grid is in peak or valley periods and adjust the weight of energy efficiency E accordingly; or dynamically adjust the weight of battery health change ΔH based on the battery's age.

[0087] After evaluating all possible paths, the system selects the path with the highest evaluation score as the optimal charging path. This process is implemented using an ant colony optimization algorithm, where each "ant" represents a possible charging path, and pheromone updates are based on the path's evaluation score. Through multiple iterations, the algorithm gradually converges to an optimal or near-optimal charging path.

[0088] The benefit of this embodiment is that by introducing a multi-dimensional evaluation function and a dynamic weight adjustment mechanism, the charging control system can adapt to different charging needs and environmental conditions more flexibly. This not only improves the overall efficiency of the charging process, but also takes into account the balance of multiple aspects such as battery life, energy utilization and safety. At the same time, the method is highly scalable and can introduce new evaluation parameters or adjust the calculation method of existing parameters as needed to adapt to new battery technologies or charging scenarios that may appear in the future. For example, the charging control method described in this embodiment can realize the intelligence and personalization of the charging process, provide electric vehicles with a better charging method, and provide grid operators and charging station managers with a more efficient charging resource allocation strategy. It can also be extended to the optimized management of the entire charging station or charging network.

[0089] Example 4

[0090] To address the low optimization efficiency of traditional charging control methods in complex scenarios, this embodiment further optimizes the charging control method by introducing a charging strategy optimization step based on a quantum annealing algorithm. This method leverages the parallelism of quantum computing and the quantum tunneling effect to effectively improve the ability to find the global optimal solution in a high-dimensional and complex solution space.

[0091] In this embodiment, the charging control method further includes the step of optimizing the charging strategy based on the quantum annealing algorithm:

[0092] Construct a qubit representation of the charging strategy.

[0093] Design a quantum Hamiltonian operator to map the charging optimization problem to the quantum system.

[0094] A quantum annealing process is performed to find the optimal charging strategy by adjusting the quantum tunneling effect.

[0095] The optimal charging strategy obtained by quantum state collapse is converted into an actual charging strategy.

[0096] Specifically, the system first constructs a qubit representation of the charging strategy. This step involves encoding the various control parameters of the charging process into a sequence of qubits. For example, n qubits can be used to represent the discrete levels of charging power, m qubits to represent the divisions of charging time, and k qubits to represent the temperature control strategy. This quantum encoding method allows the system to conduct parallel searches in a 2n+m+k-dimensional solution space, greatly expanding the space of explorable strategies.

[0097] Furthermore, designing a quantum Hamiltonian involves mapping the charging optimization problem to a quantum system. The Hamiltonian H can be expressed as a linear combination of the problem Hamiltonian HP and the driving Hamiltonian HD: H(t) = A(t)HP + B(t)HD. HP encodes the objective function and constraints of the charging optimization problem, such as minimizing charging time or maximizing battery life; HD is used to introduce quantum fluctuations, typically in the form of a transverse magnetic field. A(t) and B(t) are time-varying coefficients used to control the quantum annealing process.

[0098] It's important to understand that during quantum annealing, the system begins in an initial superposition state (typically the HD ground state) and gradually evolves to the HP ground state by gradually decreasing B(t) and increasing A(t). This process can be simulated by solving the Schrödinger equation or using Monte Carlo methods. Quantum tunneling allows the system to traverse energy barriers that are classically impossible, potentially allowing it to escape from local optima and find the global optimal solution.

[0099] Understandably, tuning the parameters of the quantum annealing process significantly impacts optimization results. Designing the annealing schedule requires a trade-off between annealing speed and solution quality. One feasible approach is to employ an adaptive annealing schedule, dynamically adjusting the rates of change of A(t) and B(t) based on the instantaneous state of the system. Furthermore, introducing quantum fluctuation-assisted annealing (QFAA) technology can further improve algorithm performance, particularly when dealing with complex constraints.

[0100] Furthermore, after the quantum annealing process is complete, the resulting quantum state needs to be collapsed back to a classical state. This step is usually achieved through measurement. The measurement result corresponds to a specific charging strategy. However, due to the probabilistic nature of quantum measurement, a single measurement may not directly yield the optimal solution. Therefore, in practice, multiple measurements are typically performed, and the result with the lowest energy (i.e., the optimal objective function value) is selected as the final charging strategy.

[0101] Furthermore, converting the optimal charging strategy derived from quantum state collapse into an actual charging strategy involves decoding the qubit sequence into specific charging parameters, such as the charging power curve and charging schedule. This process also requires consideration of the actual hardware limitations and safety margins, and may require some post-processing, such as smoothing the power curve and adjusting the charging time to match the requirements of the battery management system.

[0102] The benefit of this embodiment lies in the fact that, by introducing the quantum annealing algorithm, the charging control system can more effectively find the global optimal solution in a complex multidimensional solution space. Compared with traditional optimization methods, the quantum annealing algorithm has greater parallelism and the ability to escape local optima, making it particularly suitable for handling complex charging optimization problems with multiple constraints and objectives. The charging control method described in this embodiment can achieve more refined and efficient charging strategy optimization.

[0103] Example 5

[0104] To address the limitations of traditional integer-order battery models in describing battery dynamic characteristics, this example further optimizes the charging control method by introducing a battery model construction step based on fractional-order calculus. This approach more accurately captures the nonlinearity and memory effects of the battery during the charging and discharging process, thereby providing more precise battery state prediction and charging strategy optimization.

[0105] In this embodiment, the charging control method further includes the step of constructing a battery model based on fractional calculus:

[0106] The dynamic characteristics of the battery are described using fractional differential equations.

[0107] The battery state equation is defined using Caputo fractional derivatives.

[0108] The order and coefficients of the fractional-order battery model are determined through parameter identification algorithm.

[0109] Predict the battery response under different charging strategies based on the fractional-order battery model.

[0110] The state equation of the fractional-order battery model is:

[0111] , where D α and D β denote the α-order and β-order fractional derivatives respectively, V(t) is the battery voltage, I(t) is the charging current, a, b, c, d are model parameters, α and β are fractional orders, 0<α, β<1.

[0112] Specifically, fractional-order differential equations are first used to describe the dynamic characteristics of batteries. Fractional-order calculus has unique advantages in dealing with systems with long-range correlations and non-Markov processes. Batteries, due to the complexity of their electrochemical reactions, precisely conform to these characteristics. By introducing fractional-order derivatives, the diffusion processes and charge transport within the battery, as well as the electrochemical reaction dynamics on the electrode surface, can be better described.

[0113] Furthermore, this embodiment uses Caputo's fractional derivatives to define the battery state equation. Caputo's fractional derivatives have excellent physical meaning and mathematical properties, making them suitable for describing systems with initial value problems, such as batteries. The state equation comprehensively considers the battery's voltage response, current rate of change, and the influence of its historical state, providing a more comprehensive description of the battery's dynamic behavior.

[0114] It should be understood that the purpose of determining the order and coefficients of the fractional-order battery model is to construct an accurate battery model. This embodiment achieves this goal through a parameter identification algorithm. Specifically, an intelligent optimization method such as a particle swarm optimization algorithm (PSO) or a genetic algorithm (GA) can be used to minimize the error between the model prediction value and the actual measurement data as the objective function, while optimizing the parameters α, β, a, b, c, and d. In order to improve the accuracy and efficiency of parameter identification, a multi-objective optimization strategy can be adopted, taking into account the fitting errors in the time domain and frequency domain at the same time.

[0115] Understandably, the parameter identification process requires high-quality experimental data. This typically involves systematic testing of the battery's response under various charge and discharge conditions, including constant current charging, pulse charging, and dynamic load testing. To capture the battery's behavior under varying operating and environmental conditions, testing should encompass temperature, state of charge (SOC), and charge and discharge rates. Furthermore, given the variability of battery performance with cycle number, the model parameter identification process should be performed regularly to accommodate battery aging characteristics.

[0116] Furthermore, based on the constructed fractional-order battery model, the battery's response under different charging strategies can be predicted. This step involves numerically solving fractional-order differential equations. Common methods include the modified Adams method and the predictor-corrector method. In practical applications, to improve computational efficiency, fast convolution algorithms or approximate methods based on frequency domain analysis can be used to accelerate the calculation of fractional-order derivatives. In this way, the impact of different charging strategies on key parameters such as battery voltage, temperature, and state of charge can be simulated in real time, providing a theoretical basis for optimizing charging strategies.

[0117] The benefit of this embodiment lies in the fact that, by introducing fractional-order calculus theory, the battery model can more accurately describe the nonlinear dynamic characteristics and memory effect of the battery. This high-precision battery model provides a more reliable foundation for optimizing charging strategies and facilitates more refined charging control. The advantages of the fractional-order battery model are particularly evident under fast charging and extreme temperature conditions, allowing for better prediction of battery response, thereby formulating safer and more efficient charging strategies.

[0118] The fractional calculus-based battery modeling method described in this example significantly improves the performance of charging control systems. This method is not only applicable to the modeling of single cells but can also be extended to the design of battery packs and battery management systems (BMSs).

[0119] Example 6

[0120] To address the inefficiency and battery performance degradation associated with conventional constant-current and constant-voltage charging methods, this embodiment further optimizes the charging control method by introducing a charging pulse technique based on chaos theory. This technique leverages the irregularities and sensitive dependencies of chaotic systems to generate dynamically changing charging pulse sequences, thereby improving charging efficiency, reducing battery damage, and extending battery life.

[0121] In this embodiment, a Lorenz chaotic system is first constructed as the basis for generating the charging pulse sequence. The Lorenz system is a classic three-dimensional nonlinear dynamic system, and its mathematical model is defined as follows:

[0122] dx / dt = σ(yx)

[0123] dy / dt = x(ρ-z)-y

[0124] dz / dt = xy-γz

[0125] Here, σ, ρ, and γ are system parameters used to control the chaotic characteristics of the system. Typical parameter choices might be σ = 10, ρ = 28, and γ = 8 / 3, but these parameters can be fine-tuned based on actual charging needs. It is important to note that the analytical solution of the Lorenz system cannot be directly obtained. Therefore, in practical applications, numerical integration methods, such as the fourth-order Runge-Kutta method, are required to solve the system trajectory.

[0126] Furthermore, chaotic charging is achieved by mapping the chaotic sequence to the amplitude and frequency of the charging pulse. This process can be accomplished by designing an appropriate mapping function. For example, the x variable can be mapped to the pulse amplitude, and the y variable can be mapped to the pulse frequency. The mapping function needs to consider the battery's charging characteristics and safety limitations to ensure that the generated pulse sequence is within the battery's safe operating range. The mapping method can be expressed as:

[0127]

[0128]

[0129] Among them, I(t) is the charging current amplitude, f(t) is the pulse frequency, I min and I max are the minimum and maximum charging currents allowed, respectively, f min and fmax are the minimum and maximum pulse frequencies allowed, respectively. x(t) and y(t) are the x and y variables of the Lorenz system. min 、x max 、y min 、y max is the value range of the x and y variables.

[0130] It's important to understand that by adjusting the parameters of the Lorenz system, charging pulse trains with different characteristics can be generated. For example, increasing the value of ρ can induce more complex chaotic behavior in the system, thereby generating a more diverse pulse train. This parameter adjustment can be performed dynamically based on factors such as the battery's charging stage, temperature conditions, and state of charge. To achieve this, a parameter adaptation mechanism can be designed to adjust the values of σ, ρ, and γ based on real-time monitoring of the battery's status.

[0131] It is understandable that real-time evaluation of charging performance and dynamic selection of the optimal chaotic charging pulse pattern are important steps in ensuring charging efficiency and safety. This can be achieved by establishing a multi-objective evaluation function that comprehensively considers factors such as charging speed, battery temperature changes, and Coulomb efficiency. The evaluation function can be expressed as:

[0132]

[0133] Among them, SOC is the current state of charge, SOC target is the target state of charge, T is the battery temperature, T opt is the optimal operating temperature, η is the charging efficiency, and k1, k2, and k3 are weight coefficients. The system continuously calculates the value of this evaluation function and dynamically adjusts the Lorenz system parameters based on the results to optimize the charging pulse pattern.

[0134] The benefit of this embodiment is that by incorporating chaos theory into the charging pulse design, the charging control system can generate a more complex and dynamic charging strategy. This approach not only effectively reduces polarization within the battery, but also promotes electrolyte convection through irregular charging pulses, improving ion transport efficiency. Furthermore, the unpredictability of chaotic pulses helps prevent the formation of a fixed stress distribution within the battery, thereby slowing down the degradation of battery performance. The chaos theory-based charging pulse method described in this embodiment enables a more intelligent and efficient battery charging process.

[0135] Example 7

[0136] To address the shortcomings of traditional charging control methods in terms of long-term optimization and dynamic adaptation, this embodiment further optimizes the charging control method by introducing a long-term charging strategy optimization step based on reinforcement learning. Through continuous interactive learning with the environment, the charging strategy is adaptively adjusted to adapt to different battery states, usage patterns, and environmental conditions, thereby achieving a balance between charging efficiency and battery life during long-term use.

[0137] In this example, the state space, action space, and reward function of the charging process are first defined. The state space can include parameters such as the battery's state of charge (SOC), state of health (SOH), temperature, voltage, and internal resistance, as well as external factors such as ambient temperature and charging time. The action space can be defined as different charging current levels or charging power curves. The design of the reward function needs to comprehensively consider multiple objectives such as charging speed, battery life, and energy efficiency, and can be expressed as:

[0138]

[0139] Where ΔSOC / Δt represents the charging rate, ΔSOH represents the change in health status, η represents the charging efficiency, T is the battery temperature, and T opt is the optimal operating temperature, and m1, m2, m3, and m4 are weight coefficients. The reward function encourages fast charging while penalizing adverse effects on battery health and excessive temperature deviations.

[0140] Furthermore, this embodiment uses the Deep Q-Network (DQN) algorithm to learn the optimal charging strategy. The DQN algorithm combines Q-learning with deep neural networks and can handle high-dimensional state spaces. The network structure can use a multi-layer perceptron (MLP), with the input layer corresponding to the state vector and the output layer corresponding to the Q value of each action. The hidden layer can use the ReLU activation function to introduce nonlinearity.

[0141] It is understandable that balancing exploration and exploitation through an ε-greedy strategy is a key issue in reinforcement learning. In this embodiment, a time-decreasing ε value can be used. Initially, a high ε value is set to encourage exploration, and as learning progresses, the ε value is gradually reduced to increase the exploitation of the learned strategy. The decay of ε can follow an exponential decay law:

[0142]

[0143] Among them, ε start and ε end are the initial and final ε values, d rate is the decay rate, and t is the number of training steps.

[0144] In this embodiment, the policy network is regularly updated to continuously optimize the charging control strategy. This can be accomplished by setting a fixed update frequency, such as updating network parameters every N charging cycles. During the update process, stochastic gradient descent (SGD) or its variants, such as the Adam optimizer, can be used to minimize the loss function. To improve learning efficiency and generalization, prioritized experience replay (PER) can be introduced to adjust the probability of sample selection based on the magnitude of the TD error.

[0145] The benefit of this embodiment is that, by introducing long-term charging strategy optimization based on reinforcement learning, the charging control system can adaptively learn and improve the charging strategy. This not only considers the effects of a single charge, but also focuses on the long-term performance and lifespan of the battery. It can dynamically adjust the charging strategy based on battery usage patterns, environmental conditions, and aging status, achieving multi-objective optimization of charging efficiency and battery life.

[0146] Example 8

[0147] To address the challenges of low charging efficiency, short battery life, and unintelligent energy management in traditional smart lock charging systems under complex usage scenarios, this embodiment further proposes an overall charging system architecture, introducing an intelligent charging control solution based on a multidimensional state space and an ant colony optimization algorithm. This system achieves comprehensive optimization and dynamic adjustment of the charging process through the collaborative operation of multiple functional modules.

[0148] As shown in FIG5 , the charging system of this embodiment includes:

[0149] The multi-dimensional state space construction module 100 is used to construct the multi-dimensional state space of the smart lock battery charging system.

[0150] The ant colony optimization path planning module 200 is used to design a charging path planning model based on the ant colony optimization algorithm, and regards the charging process as a process in which ants search for the optimal path in a multi-dimensional state space.

[0151] The real-time data acquisition module 300 is used to collect the working status information and environmental parameters of the smart lock in real time.

[0152] The deep neural network prediction module 400 is used to receive the collected information and predict the usage pattern and energy demand of the smart lock.

[0153] The parameter dynamic adjustment module 500 is used to dynamically adjust the parameters in the ant colony optimization algorithm based on the prediction results.

[0154] The optimization algorithm execution module 600 is used to execute the ant colony optimization algorithm in the adjusted multi-dimensional state space to find the optimal charging path.

[0155] The charging strategy conversion module 700 is used to convert the optimal charging path into a specific charging control strategy.

[0156] The charging control execution module 800 is used to execute the charging control strategy in real time.

[0157] The closed-loop feedback monitoring module 900 is used to continuously monitor the charging effect and provide feedback.

[0158] In this embodiment, the multi-dimensional state space construction module 100 first comprehensively models the state of the smart lock battery charging system. This state space may include internal parameters such as the battery's state of charge (SOC), state of health (SOH), temperature, voltage, internal resistance, and external factors such as ambient temperature, humidity, frequency of use, and time. Each factor can be regarded as a dimension in the state space, together forming a high-dimensional space. In order to effectively process this high-dimensional data, dimensionality reduction techniques such as principal component analysis (PCA) or autoencoders can be used to map the original high-dimensional space to a more compact low-dimensional representation to improve the efficiency of subsequent optimization algorithms.

[0159] Furthermore, the ant colony optimization path planning module 200 designs a highly adaptable charging path planning model based on the constructed multidimensional state space. In this model, the charging process is abstracted as an ant's search for an optimal path in the multidimensional state space. Each ant represents a possible charging strategy, and its path corresponds to the sequence of transitions from the initial state to the target charging state.

[0160] It's important to understand that the real-time data acquisition module 300 acts as an information provider throughout the system. It collects real-time information about the smart lock's operating status and environmental parameters through an embedded sensor network. This may include high-frequency sampling of battery voltage, current, and temperature, as well as information such as ambient temperature and humidity, door lock opening and closing times, and user operation patterns. To ensure data reliability and real-time performance, a distributed data acquisition architecture can be employed, combined with edge computing technology for preliminary data cleaning and compression.

[0161] It is understood that after receiving the collected real-time data, the deep neural network prediction module 400 needs to predict the future usage patterns and energy requirements of the smart lock. This can be achieved by using a time series prediction model such as a long short-term memory (LSTM) network or a temporal convolutional network (TCN). The network input can be usage data and environmental parameters from the past, and the output is the predicted usage frequency and energy consumption for the future. The model can be trained using a sliding window method, regularly updating the model parameters with the latest data to adapt to dynamic changes in user behavior and the environment.

[0162] Furthermore, the dynamic parameter adjustment module 500 adjusts key parameters in the ant colony optimization algorithm in real time based on the prediction results of the deep neural network. These parameters may include the pheromone importance factor, heuristic factor, pheromone volatility coefficient, etc. The adjustment strategy can use a fuzzy logic controller to dynamically balance the algorithm's global search capability and local optimization capabilities based on predicted usage intensity and energy demand. For example, when high future usage frequency is predicted, the pheromone importance factor can be increased to enhance the utilization of historical high-quality routes.

[0163] The optimization algorithm execution module 600 executes the ant colony optimization algorithm in a multidimensional state space under the adjusted parameter settings. In each iteration, the algorithm generates a set of candidate charging paths and updates the pheromone distribution based on the path quality. Path quality assessment can comprehensively consider multiple indicators such as charging speed, battery life impact, and energy efficiency. To improve search efficiency, local search strategies, such as the 2-opt or 3-opt algorithms, can be introduced to locally optimize the best paths found.

[0164] Furthermore, the charging strategy conversion module 700 converts the optimal path found by the ant colony algorithm into a specific charging control strategy. This process involves mapping the abstract state transition sequence to actual charging parameters, such as charging current and voltage curves. This conversion can be achieved using a pre-established lookup table or by training an additional neural network. The generated charging strategy must consider hardware limitations and safety margins to ensure the safety and reliability of the charging process.

[0165] Furthermore, the charging control execution module 800 is responsible for executing the converted charging strategy in real time. This requires a high-precision, fast-response power management system capable of precisely controlling charging current and voltage. This can be achieved by using a digitally controlled DC-DC converter combined with pulse-width modulation (PWM) technology to achieve precise control of charging parameters. Furthermore, protection mechanisms such as overvoltage, overcurrent, and overtemperature protection are also required to address abnormal situations.

[0166] Furthermore, the closed-loop feedback monitoring module 900 continuously monitors the actual performance of the charging process and feeds the results back to other modules in the system. This includes real-time tracking of changes in battery status, assessing charging efficiency and its impact on battery life. This feedback information can be used to adjust the prediction model of the deep neural network, optimize the parameter settings of the ant colony algorithm, and even modify the construction method of the multidimensional state space. Through this closed-loop feedback mechanism, the system can continuously learn and improve, adapting to changes in battery characteristics and the dynamic nature of the usage environment.

[0167] The benefit of this embodiment lies in the fact that by introducing a multidimensional state space and ant colony optimization algorithm, combined with deep learning prediction and real-time parameter adjustment, a highly intelligent and adaptive charging system is achieved. This system can dynamically optimize the charging strategy based on the actual usage and environmental conditions of the smart lock, effectively extending battery life while improving charging efficiency.

[0168] Example 9

[0169] To address the limitations of traditional smart locks in energy management, environmental adaptability, and functional scalability, this embodiment further optimizes the overall design of the smart lock and introduces multiple advanced technologies and functional modules. These improvements aim to improve the energy efficiency, self-sustainability, and flexibility of the smart lock, enabling it to better adapt to complex and changing usage environments and user needs.

[0170] In this embodiment, the smart lock integrates a micro supercapacitor module to manage peak loads and achieve fast response. This supercapacitor can be made of graphene-based materials, which have high power density and fast charging and discharging characteristics. Its capacity can be designed in the range of 1-10 F, and the operating voltage is 2.7-3.0 V, which is sufficient to support the instantaneous high power requirements of the lock, such as the sudden operation of the motor drive and wireless communication module. The supercapacitor is connected in parallel with the main battery and coordinates its work through an intelligent power management circuit. In daily use, the supercapacitor is mainly responsible for handling short-term high power demands, while the main battery provides a continuous low-power supply. This configuration can significantly reduce the instantaneous high current impact on the main battery and extend battery life.

[0171] Furthermore, smart locks integrate an ambient energy harvesting system, including flexible solar cells and piezoelectric materials. Flexible solar cells can be made of perovskite materials, boasting efficiencies exceeding 20%. They can be bent and adhered to the outer surface of the lock, utilizing ambient light for continuous charging. Piezoelectric materials can be integrated into the lock's mechanical moving parts, such as buttons or shafts, to convert mechanical energy generated during user operation into electrical energy. The outputs of these two energy harvesting methods are regulated and integrated by an energy management integrated circuit, providing auxiliary power to the system and reducing reliance on the main battery.

[0172] It is important to understand that the introduction of a graphene-based heat dissipation structure improves the thermal management efficiency of smart locks. This heat dissipation structure can be made of graphene composite materials with a thermal conductivity of up to 2000-3000 W / (m·K). The heat dissipation structure is designed as a multi-layer network and is embedded around key heat sources inside the lock, such as the processor, power management unit, and charging module. Through efficient heat diffusion and heat dissipation, the operating temperature of the system can be controlled at a low level, usually not exceeding 10°C from the ambient temperature, which not only improves the reliability and life of electronic components, but also indirectly improves the charging and dissipation efficiency of the battery.

[0173] As you can imagine, the programmable charging interface design provides smart locks with greater compatibility and functional scalability. This interface utilizes the USB Type-C standard and supports multiple charging protocols, including USB, PD, and QC. The interface's control logic is managed by a programmable MCU (such as the STM32L4 series), and firmware updates can be used to support new charging standards. Furthermore, this interface supports bidirectional charging, allowing the smart lock to function as a power bank for other devices when needed. The interface's power management utilizes a dedicated power path management IC (such as the TPS2113A), which intelligently switches between internal and external power sources to ensure safe and efficient charging and discharging.

[0174] The benefits of this embodiment lie in improving the smart lock's energy efficiency and environmental adaptability by integrating advanced energy and thermal management technologies. The introduction of supercapacitors and an ambient energy harvesting system reduces reliance on the main battery and extends the device's operating time. The graphene heat dissipation structure ensures stable system operation in various environments. The programmable charging port provides greater compatibility and functional scalability.

[0175] Example 10

[0176] To address the portability and compatibility issues of the charging control method in practical applications, this embodiment further optimizes the software implementation of the charging control method, encapsulating it as a computer program executable on multiple computing platforms and storing it on a computer-readable storage medium. This implementation allows the charging control method to be more flexibly deployed in various smart devices and systems, increasing its applicability and practical value.

[0177] In this embodiment, the computer-readable storage medium can be a non-volatile storage device, such as a solid-state drive (SSD), an embedded multimedia card (eMMC), or NAND flash memory. The storage capacity can be selected based on program complexity and data requirements. The storage medium uses highly reliable industrial-grade specifications to ensure data integrity and long-term stability in harsh environments. To enhance data security, an encrypted file system can be used to encrypt stored programs and data.

[0178] The benefit of this embodiment is that by implementing the charging control method as a computer program stored on a computer-readable storage medium, the portability and applicability of the method are greatly improved. This implementation allows the charging control algorithm to be easily deployed in a variety of smart devices, from small IoT devices to large-scale energy management systems.

[0179] The computer program implementation described in this embodiment not only makes the charging control method easier to integrate and maintain, but also facilitates future functional expansion and algorithm optimization. This software-based approach lays the foundation for the continuous iteration and upgrade of intelligent charging systems, helping to promote the application of charging technology in various fields.

[0180] The above embodiments further illustrate the objectives, technical solutions and beneficial effects of the present invention. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solutions of the present invention should be included in the scope of protection of the present invention.

Claims

1. A charging control method, applied to a charging system, characterized in that: The steps of the charging control method include: Constructing a multidimensional state space of the charging system; A charging path planning model is designed based on the ant colony optimization algorithm, which regards the charging process as a process in which ants search for the optimal path in the multidimensional state space. Real-time collection of operating status information and environmental parameters of the charging system; Inputting the collected information into a pre-trained deep neural network to predict usage patterns and energy requirements of the charging system; Dynamically adjusting parameters in an ant colony optimization algorithm based on the prediction results, and executing the ant colony optimization algorithm in the adjusted multidimensional state space to find an optimal charging path; Converting the optimal charging path into a specific charging control strategy, executing the charging control strategy in real time, and continuously monitoring the charging effect; The step of designing a charging path planning model based on an ant colony optimization algorithm includes: Define a path evaluation function F(path), wherein the evaluation function F(path) includes the following evaluation parameters: charging time T, change in battery health state ΔH, charging energy efficiency E, temperature change ΔTemp, and weight coefficients w1, w2, w3, and w4; For each possible charging path, calculating the charging time T, evaluating the battery health state change ΔH, calculating the charging energy efficiency E, and measuring the temperature change ΔTemp; By formula Calculate assessment scores; Dynamically adjust the weight coefficient according to different charging scenarios and priorities; Selecting the path with the highest evaluation score as the optimal charging path; The charging control method further includes a step of optimizing the charging strategy based on a quantum annealing algorithm: Constructing a qubit representation of the charging strategy; Design a quantum Hamiltonian operator to map the charging optimization problem to the quantum system; Perform quantum annealing to find the optimal charging strategy by adjusting the quantum tunneling effect; Converting the optimal charging strategy obtained by quantum state collapse into an actual charging strategy; The charging control method further includes the step of constructing a battery model based on fractional calculus: Use fractional differential equations to describe the dynamic characteristics of batteries; The battery state equation is defined using Caputo fractional derivatives; Determine the order and coefficients of the fractional-order battery model through a parameter identification algorithm; Predicting the response of the battery under different charging strategies based on the fractional-order battery model; The state equation of the fractional-order battery model is: , where D α and D β denote the α-order and β-order fractional derivatives, V(t) is the battery voltage, I(t) is the charging current, a, b, c, d are model parameters, α and β are fractional orders, 0<α, β<1; The charging control method further comprises the step of charging pulse based on chaos theory: Construct a Lorenz chaotic system; Mapping chaotic sequences to the amplitude and frequency of charging pulses; By adjusting the parameters of the Lorenz system, charging pulse sequences with different characteristics are generated; Real-time evaluation of charging effects and dynamic selection of the optimal chaotic charging pulse mode; The Lorenz chaotic system is defined as follows: dx / dt=σ(yx) dy / dt=x(ρ-z)-y dz / dt=xy-γz Among them, σ, ρ and γ are system parameters used to change the chaotic characteristics.

2. The charging control method according to claim 1, wherein: The multidimensional state space includes at least one of battery power, charging current, charging voltage, ambient temperature, usage frequency, unlocking times, network connection status, data transmission frequency, electromagnetic interference intensity and battery health status.

3. The charging control method according to claim 1, wherein: The charging control method further includes the step of optimizing a long-term charging strategy based on reinforcement learning: Define the state space, action space, and reward function of the charging process; Use the Deep Q Network (DQN) algorithm to learn the optimal charging strategy; Use experience replay and target network techniques to improve learning stability; Use ε-greedy strategy to balance exploration and exploitation; The strategy network is updated regularly to continuously optimize the charging control strategy.

4. A charging system for implementing the charging control method according to any one of claims 1 to 3, characterized in that: The charging system comprises: Multi-dimensional state space construction module, used to construct the multi-dimensional state space of the smart lock battery charging system; The ant colony optimization path planning module is used to design a charging path planning model based on the ant colony optimization algorithm, which regards the charging process as a process in which ants search for the optimal path in a multi-dimensional state space; Real-time data acquisition module, used to collect the working status information and environmental parameters of the smart lock in real time; A deep neural network prediction module that receives the collected information and predicts the usage pattern and energy demand of the smart lock; Parameter dynamic adjustment module, used to dynamically adjust the parameters in the ant colony optimization algorithm based on the prediction results; An optimization algorithm execution module is used to execute an ant colony optimization algorithm in the adjusted multi-dimensional state space to find the optimal charging path; A charging strategy conversion module is used to convert the optimal charging path into a specific charging control strategy; Charging control execution module, used to execute charging control strategy in real time; Closed-loop feedback monitoring module, used to continuously monitor charging results and provide feedback.

5. A smart lock comprising the charging system according to claim 4, characterized in that: The smart lock also includes: Integrated micro-supercapacitor for peak load management and fast response; Flexible solar cells and piezoelectric materials for ambient energy harvesting; Graphene-based heat dissipation structure for improving thermal management efficiency; Programmable charging interface to support multiple charging protocols and bidirectional charging function.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the charging control method according to any one of claims 1 to 3 are implemented.

Images