An intelligent anti-interference wireless charging school bus control system and method based on AI large model analysis

Abstract

The present invention proposes an intelligent anti-interference wireless charging school bus control system and method based on AI large model analysis, which belongs to the field of wireless charging control, including: a wireless charging control unit, which is configured to use a constant voltage filter cascade self-anti-interference control strategy to wirelessly charge the school bus; a data acquisition and transmission unit, which is configured to collect school bus operation data in real time, and transmit the school bus operation data to a central data processing platform for data processing; an artificial intelligence analysis unit, which is configured to perform multi-dimensional analysis on the school bus operation data after data processing based on the AI ​​large model, and generate an intelligent scheduling strategy; a cloud application interface, which is configured to transmit the intelligent scheduling strategy and the school bus operation data after data processing to a mobile terminal, and the operator performs real-time query and use through the mobile terminal; an intelligent scheduling execution unit, which is configured to transmit the intelligent scheduling strategy to the school bus driver, and the school bus driver allocates resources according to the intelligent scheduling strategy. This application effectively suppresses interference and fluctuations during the charging process, and performs intelligent analysis on the school bus operation data, realizes real-time dynamic scheduling, improves the school bus operation efficiency, and optimizes resource allocation.

Description

Technical Field

[0001] The present invention relates to the field of wireless charging control technology, and in particular to an intelligent anti-interference wireless charging school bus control system and method based on AI large model analysis. Background Art

[0002] With the development and application of science and technology, and the continuous expansion of university campuses in recent years, it is necessary to deploy school buses in universities to improve the convenience of travel on campus. However, for existing campus school buses, some peak hours cannot be allocated in time, resulting in congestion in some parts of the campus, so reasonable and efficient resource allocation is necessary; in addition, most existing school buses are charged by direct charging, and the use of wireless charging can greatly improve the charging efficiency of school buses.

[0003] However, in the prior art, the existing wireless charging system is susceptible to external electromagnetic interference during the charging process, which affects the charging efficiency and stability. Since there are a large number of electronic devices and wireless signals on campus, the impact on school buses is particularly obvious. In addition, the school bus dispatching system often operates in a fixed route and timetable manner, which is not flexible enough to deal with sudden demands during peak hours, which can easily lead to a shortage of school buses in some areas, while vehicles in other areas are idle, resulting in waste of resources and traffic congestion.

[0004] Therefore, finding a method that can not only wirelessly charge school buses and improve the anti-interference ability during the charging process, but also scientifically and efficiently deploy school buses is a technical problem that needs to be urgently solved by technical personnel in this field. Summary of the invention

[0005] In view of this, the present invention proposes an intelligent anti-interference wireless charging school bus control system and method based on AI large model analysis, which can effectively suppress interference and fluctuations in the charging process, and perform intelligent analysis of school bus operation data to achieve real-time dynamic scheduling, thereby improving school bus operation efficiency, optimizing resource allocation, and improving service quality.

[0006] The technical solution of the present invention is achieved in this way:

[0007] In a first aspect, the present invention provides an intelligent anti-interference wireless charging school bus control system based on AI large model analysis, comprising:

[0008] A wireless charging control unit, which is configured to include a primary circuit, a secondary circuit and a constant voltage filter cascaded auto-disturbance rejection control strategy, and uses the constant voltage filter cascaded auto-disturbance rejection control strategy to wirelessly charge the school bus, wherein the constant voltage filter cascaded auto-disturbance rejection control strategy includes a first-stage auto-disturbance rejection controller, a second-stage auto-disturbance rejection controller and a Butterworth filter;

[0009] A data collection and transmission unit configured to collect school bus operation data in real time and transmit the school bus operation data to a central data processing platform for data processing;

[0010] An artificial intelligence analysis unit, which is configured to perform multi-dimensional analysis on the processed school bus operation data based on the AI ​​big model and generate an intelligent dispatching strategy;

[0011] A cloud application interface is configured to transmit the intelligent dispatching strategy and the processed school bus operation data to the mobile terminal, and the operator can query and use it in real time through the mobile terminal;

[0012] The intelligent scheduling execution unit is configured to transmit the intelligent scheduling strategy to the school bus driver, and the school bus driver allocates resources according to the intelligent scheduling strategy.

[0013] On the basis of the above technical solution, preferably, the constant voltage filter cascaded anti-disturbance control strategy specifically includes:

[0014] The first-stage anti-disturbance controller processes the output voltage of the DC-DC circuit in the secondary circuit after passing through the Butterworth filter; the secondary circuit includes a rectifier output circuit and a DC-DC circuit;

[0015] The second-stage ADRC processes the output voltage of the secondary-side DC-DC circuit;

[0016] The Butterworth filter, the first-stage ADRC, the second-stage ADRC and the secondary-side DC-DC circuit are combined into a new composite cascade system to obtain a constant voltage filter cascade ADRC control strategy.

[0017] On the basis of the above technical solution, preferably, the first-level active disturbance rejection controller includes a first-level linear extended state observer, a first state feedback controller, a first disturbance compensator and a first control rate generator, wherein:

[0018] The two ends of the first-stage linear extended state observer are connected to the Butterworth filter and the first state feedback controller respectively, the other end of the first state feedback controller is connected to the first end of the first control rate generator, the second end of the first control rate generator is connected to the first disturbance compensator, and the first disturbance compensator is also connected to the second disturbance compensator. The calculation formula of the first-stage active disturbance rejection controller is:

[0019]

[0020]

[0021]

[0022] in, represents the change of state variables in the first-level linear extended state observer, A represents the state matrix, B represents the input matrix, B = [0 b 0 0] T , b 0 represents the gain of the second-stage ADRC, L represents the bandwidth parameter, and y q represents the input of the Butterworth filter, represents the observed value of the first-stage ADRC, C 1 Indicates that z 1 Mapped to the output of the secondary circuit, z 1 represents the state vector of the first-level linear extended state observer, represents the acceleration output by the Butterworth filter, d q represents the total disturbance of the Butterworth filter, b 1 represents the control gain of the wireless charging unit, represents the output of the first control rate generator, and n is the noise interference.

[0023] On the basis of the above technical solution, preferably, the second-level active disturbance rejection controller includes a second-level linear extended state observer, a second state feedback controller, a second disturbance compensator and a second control rate generator, wherein:

[0024] The two ends of the second-stage linear extended state observer are respectively connected to the DC-DC circuit of the secondary circuit and the second state feedback controller, the other end of the second state feedback controller is connected to the first end of the second control rate generator, the second end of the second control rate generator is connected to the second disturbance compensator, the second disturbance compensator is connected to the secondary circuit through PWM, and the calculation formula of the second-stage active disturbance rejection controller is:

[0025]

[0026]

[0027] in, represents the change of state variables in the second-level linear extended state observer, z 2 represents the state vector of the second-level linear extended state observer, p f represents the output noise of the secondary circuit, represents the observed value of the second-level ADRC, represents the output of the second control rate generator, C 2 Indicates that z 2 Mapped to the output of the secondary circuit, y 2 Represents the actual output of the second-stage ADRC.

[0028] On the basis of the above technical solution, preferably, the calculation formula of the constant voltage filter cascade auto-disturbance rejection control strategy is:

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035] in, represents the rate of change of z, z represents the state vector of the constant voltage filter cascade auto-disturbance rejection control strategy, u represents the input of the constant voltage filter cascade auto-disturbance rejection control strategy, y represents the actual output of the constant voltage filter cascade auto-disturbance rejection control strategy, represents the estimated output of the constant voltage filter cascaded ADRC strategy, k p represents the proportional gain parameter of the first-stage auto-winding controller, Indicates z 1 The state variable after the Butterworth filter, z' 1 、z' 2 and z' 3 Represent the three components of z, k 2 represents the differential gain parameter of the first-stage auto-winding controller, k 3 represents the disturbance compensation gain parameter of the first-stage auto-winding controller, Indicates z q The time derivative of q represents the state matrix of the cascade system, where the cascade system includes the first-stage ADRC and the second-stage ADRC, B q represents the input matrix of the cascade system, u q represents the control input of the cascade system, L 1 represents the observer gain matrix of the cascade system, represents the estimated output of the cascade system, y q represents the actual output of the cascade system, z q represents the state vector of the cascade system, b 1 represents the control gain of the cascade system, represents the proportional gain parameter of the cascade system, v r represents the reference input, and Indicates z q The three components of k4 represents the differential gain parameter of the cascade system, k 5 Denotes the disturbance compensation gain parameter of the cascade system

[0036] Further preferably, the school bus operation data includes card swiping data and vehicle data, wherein the card swiping data includes the card swiping time, card swiping location and card swiping type, and the vehicle data includes the number of people in the bus, the distance between the previous / rear school bus, the number of people in the previous / rear school bus, the recent average number of people getting on the bus at the alighting stop in different time periods, and the number of people getting on the bus at the previous school alighting stop.

[0037] On the basis of the above technical solution, preferably, the multi-dimensional analysis includes:

[0038] The processed school bus operation data is transferred to a high-dimensional feature space through mapping to extract key features of the data;

[0039] Set up analysis dimensions and create dimension prompts for each analysis dimension;

[0040] Use AI big models to conduct in-depth analysis of key data features and dimensional prompts, and summarize the analysis results of each dimension to obtain the current school bus operation analysis results.

[0041] In a second aspect, the present invention provides an intelligent anti-interference control method for wireless charging school bus based on AI large model analysis, which is applied to the intelligent anti-interference wireless charging school bus control system as described above, comprising the following steps:

[0042] S1, taking the output voltage during charging as the control variable and setting the voltage as the input variable;

[0043] S2. Design the first-stage auto-disturbance rejection controller and the second-stage auto-disturbance rejection controller, and combine the first-stage Butterworth filter, the constant voltage filter cascade auto-disturbance rejection controller and the secondary side DC-DC circuit of the wireless charging unit into a constant voltage filter cascade auto-disturbance rejection control strategy;

[0044] S3, using constant voltage filter cascaded anti-disturbance control strategy to wirelessly charge the school bus;

[0045] S4, collect school bus operation data in real time, and transmit the school bus operation data to the central data processing platform for data processing;

[0046] S5. Perform multi-dimensional analysis on the processed school bus operation data based on the AI ​​big model and generate intelligent scheduling strategies;

[0047] S6, transmitting the intelligent dispatching strategy and the processed school bus operation data to the mobile terminal, and the operator conducts real-time query and use through the mobile terminal;

[0048] S7. Transmit the intelligent scheduling strategy to the school bus driver, and the school bus driver allocates resources according to the intelligent scheduling strategy.

[0049] In a third aspect, the present invention provides an electronic device, comprising: at least one processor, at least one memory, a communication interface and a bus; wherein:

[0050] The processor, memory, and communication interface communicate with each other via the bus;

[0051] The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement the intelligent anti-interference wireless charging school bus control system as described above.

[0052] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to implement the intelligent anti-interference wireless charging school bus control system as described above.

[0053] The intelligent anti-interference wireless charging school bus control system of the present invention has the following beneficial effects compared with the prior art:

[0054] (1) The constant voltage filter cascaded anti-disturbance control strategy is used to effectively suppress interference and fluctuations during the charging process, and the first-level anti-disturbance control controller and the second-level anti-disturbance control controller are used to improve the robustness of wireless charging and charging efficiency. The school bus operation data is intelligently analyzed to achieve real-time dynamic scheduling, improve school bus operation efficiency, optimize resource allocation, and improve service quality.

[0055] (2) By using the constant voltage filter cascade anti-disturbance control strategy to monitor the input voltage, output voltage, charging current and other parameters in real time, the stability of the output voltage is ensured, and a constant and reliable charging voltage is provided for the school bus. The Butterworth filter is used to effectively filter out high-frequency noise, improve the anti-interference ability of the wireless charging unit, and reduce the interruption or efficiency reduction of the charging process. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 It is a block diagram of the intelligent anti-interference wireless charging school bus control system of the present invention;

[0058] Figure 2It is a structural block diagram of the first-level anti-interference controller of the intelligent anti-interference wireless charging school bus control system of the present invention;

[0059] Figure 3 A school bus flow chart of the intelligent anti-interference wireless charging school bus control method of the present invention;

[0060] Figure 4 This is a flow chart of the intelligent anti-interference wireless charging school bus control method of the present invention;

[0061] Figure 5 It is a flowchart of the intelligent anti-interference wireless charging school bus control method of the present invention;

[0062] Figure 6 It is a comparative simulation output diagram of the intelligent anti-interference wireless charging school bus control method of the present invention. DETAILED DESCRIPTION

[0063] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] like Figure 1 As shown, the present invention provides an intelligent anti-interference wireless charging school bus control system based on AI large model analysis, including:

[0065] A wireless charging control unit is configured to include a primary circuit, a secondary circuit and a constant voltage filter cascade auto-disturbance rejection control strategy, and a constant voltage filter cascade auto-disturbance rejection control (FCLADRC) strategy is used to wirelessly charge a school bus, wherein the constant voltage filter cascade auto-disturbance rejection control (FCLADRC) strategy includes a first-stage auto-disturbance rejection controller, a second-stage auto-disturbance rejection controller and a Butterworth filter.

[0066] The primary circuit receives AC power and converts it into high-frequency AC power. The high-frequency AC power generates an alternating magnetic field through the primary coil. After the secondary circuit receives the induced AC power, it converts it into DC power through the rectifier circuit. The rectified and filtered DC power passes through the DC-DC conversion circuit to adjust the voltage to the level required by the school bus. The adjusted DC power is used to power the school bus battery or motor system through the constant voltage filter cascade self-disturbance rejection control strategy.

[0067] The constant voltage filter cascade self-anti-disturbance control system of the present application can not only improve the charging response time of the system, but also effectively suppress external interference. When the rated voltage changes, it can also restore the set voltage more quickly. The school bus can be wirelessly charged when it reaches the designated location. The wireless charging is sufficient to meet the operation of the school bus. At the same time, the electric energy received by the wireless charging can also supply the operation of the school bus card swiping system.

[0068] In an embodiment of the present application, the constant voltage filter cascaded active disturbance rejection control strategy specifically includes:

[0069] The first-stage anti-disturbance controller processes the output voltage of the DC-DC circuit in the secondary circuit after passing through the Butterworth filter; the secondary circuit includes a rectifier output circuit and a DC-DC circuit;

[0070] The second-stage ADRC processes the output voltage of the secondary-side DC-DC circuit;

[0071] The Butterworth filter, the first-stage ADRC, the second-stage ADRC and the secondary-side DC-DC circuit are combined into a new composite cascade system to obtain a constant voltage filter cascade ADRC control strategy.

[0072] The first-stage anti-disturbance controller receives the secondary output voltage processed by the Butterworth filter and generates a control signal to adjust the primary circuit parameters. The second-stage anti-disturbance controller receives the output voltage of the DC-DC circuit and generates a control signal to adjust the DC-DC circuit parameters. The Butterworth filter is used to filter out the high-frequency noise of the secondary circuit to enhance the anti-interference ability.

[0073] It can be understood that the constant voltage filter cascade self-disturbance rejection control of the present application not only filters the system output ripple, but also solves the amplitude and phase loss caused by the use of filters, effectively improving various aspects of the control system performance.

[0074] By utilizing the constant voltage filter cascade self-anti-disturbance control strategy to monitor the input voltage, output voltage, charging current and other parameters in real time, the parameters of the primary and secondary circuits are dynamically adjusted according to the monitoring results to ensure the stability of the output voltage and provide a constant and reliable charging voltage for the school bus. The Butterworth filter is used to effectively filter out high-frequency noise, improve the anti-interference ability of the wireless charging unit, and reduce interruptions or reduced efficiency during the charging process.

[0075] Furthermore, the first-stage ADRC includes a first-stage linear extended state observer, a first state feedback controller, a first disturbance compensator and a first control rate generator, wherein:

[0076] The two ends of the first-stage linear extended state observer are connected to the Butterworth filter and the first state feedback controller respectively, the other end of the first state feedback controller is connected to the first end of the first control rate generator, the second end of the first control rate generator is connected to the first disturbance compensator, and the first disturbance compensator is also connected to the second disturbance compensator. The calculation formula of the first-stage active disturbance rejection controller is:

[0077]

[0078]

[0079]

[0080] in, represents the change of state variables in the first-level linear extended state observer, A represents the state matrix, B represents the input matrix, B = [0 b 0 0] T , b 0 represents the gain of the second-stage ADRC, L represents the bandwidth parameter, and y q represents the input of the Butterworth filter, represents the observed value of the first-stage ADRC, C 1 Indicates that z 1 Mapped to the output of the secondary circuit, z 1 represents the state vector of the first-level linear extended state observer, represents the acceleration output by the Butterworth filter, d q represents the total disturbance of the Butterworth filter, b 1 represents the control gain of the wireless charging unit, represents the output of the first control rate generator, and n is the noise interference.

[0081] Furthermore, the second-stage ADRC includes a second-stage linear extended state observer, a second state feedback controller, a second disturbance compensator and a second control rate generator, wherein:

[0082] The two ends of the second-stage linear extended state observer are respectively connected to the DC-DC circuit of the secondary circuit and the second state feedback controller, the other end of the second state feedback controller is connected to the first end of the second control rate generator, the second end of the second control rate generator is connected to the second disturbance compensator, the second disturbance compensator is connected to the secondary circuit through PWM, and the calculation formula of the second-stage active disturbance rejection controller is:

[0083]

[0084]

[0085] in, represents the change of state variables in the second-level linear extended state observer, z 2 represents the state vector of the second-level linear extended state observer, p f represents the output noise of the secondary circuit, represents the observed value of the second-level ADRC, represents the output of the second control rate generator, C 2 Indicates that z 2 Mapped to the output of the secondary circuit, y 2 Represents the actual output of the second-stage ADRC.

[0086] like Figure 2 As shown, Figure 2 The first-stage ADRC also includes a linear tracking differentiator (LTD) and a linear state error feedback (LSEF). LTD is used to generate a smooth reference signal. LSEF uses the output of LESO to calculate the control law. The control signal is normalized by 1 / b_0 and acts on the controlled object. At the same time, the external disturbance also acts on the controlled object, so that the first-stage ADRC can estimate and compensate for the disturbance in real time, thereby improving the anti-interference ability and control accuracy of the system.

[0087] In one embodiment of the present application, the structural block diagram of the first-stage active disturbance rejection controller is the same as the structural block diagram of the second-stage active disturbance rejection controller.

[0088] In one embodiment of the present application, the calculation formula of the first control rate generator is:

[0089] u 0 =k p (v r -z 1 )-k d z 2

[0090] Among them, u 0 represents the output signal of the first control rate generator, k p represents the proportional gain parameter of the first control rate generator, v r represents the reference value, k d represents the differential gain coefficient of the first control rate generator.

[0091] It can be understood that the second control rate generator is the same as the first control rate generator, and thus will not be described in detail.

[0092] In one embodiment of the present application, the output variable of the Butterworth filter is defined as x q =y q, the Butterworth filter and the secondary circuit constitute the first composite cascade system, and the calculation formula of the first composite cascade system is:

[0093]

[0094]

[0095]

[0096] y q =x q

[0097] Among them, x q represents the output of the Butterworth filter, f q represents a nonlinear function, Represents x q The acceleration of Represents x q speed, y represents the output number of the first compound cascade system, represents the total disturbance of the first composite cascade system, u represents the input of the first composite cascade system, represents the observation value of the Butterworth filter, y q Represents the output of the first composite cascade system after filtering.

[0098] make Then the expanded state space equation of the first composite cascade system is:

[0099]

[0100]

[0101] y=Cx+n

[0102] y q =Cx q

[0103] Among them, A q represents the state matrix of the cascade system, where the cascade system includes the first-stage ADRC and the second-stage ADRC, B q represents the input matrix of the cascade system, S q represents the disturbance input matrix of the cascade system, H represents the disturbance of the Butterworth filter, It represents the rate of change of the state variable of the secondary circuit, P represents the disturbance of the secondary circuit, and S represents the impact of the disturbance on the secondary circuit.

[0104] Furthermore, the calculation formula of the constant voltage filter cascade anti-disturbance control strategy is:

[0105]

[0106]

[0107]

[0108]

[0109]

[0110]

[0111] in, represents the rate of change of z, z represents the state vector of the constant voltage filter cascade auto-disturbance rejection control strategy, u represents the input of the constant voltage filter cascade auto-disturbance rejection control strategy, y represents the actual output of the constant voltage filter cascade auto-disturbance rejection control strategy, represents the estimated output of the constant voltage filter cascaded ADRC strategy, k p represents the proportional gain parameter of the first-stage auto-winding controller, Indicates z 1 The state variable after the Butterworth filter, z' 1 、z' 2 and z' 3 Represent the three components of z, k 2 represents the differential gain parameter of the first-stage auto-winding controller, k 3 represents the disturbance compensation gain parameter of the first-stage auto-winding controller, Indicates z q The time derivative of q represents the state matrix of the cascade system, where the cascade system includes the first-stage ADRC and the second-stage ADRC, B q represents the input matrix of the cascade system, u q represents the control input of the cascade system, L 1 represents the observer gain matrix of the cascade system, represents the estimated output of the cascade system, y q represents the actual output of the cascade system, z q represents the state vector of the cascade system, b 1 represents the control gain of the cascade system, represents the proportional gain parameter of the cascade system, v r represents the reference input, and Indicates z q The three components of k 4 represents the differential gain parameter of the cascade system, k 5 Represents the disturbance compensation gain parameter of the cascade system.

[0112] A data acquisition and transmission unit is configured to collect school bus operation data in real time and transmit the school bus operation data to a central data processing platform for data processing; wherein the school bus operation data includes card swiping data and vehicle data, wherein the card swiping data includes card swiping time, card swiping location and card swiping type, and the vehicle data includes the number of people in the car, the distance between the previous / rear school bus, the number of people in the previous / rear school bus, the recent average number of people getting on the bus at the alighting stop in different time periods, and the number of people getting on the bus at the previous school alighting stop.

[0113] The artificial intelligence analysis unit is configured to perform multi-dimensional analysis on the school bus operation data after data processing based on the AI ​​big model and generate an intelligent scheduling strategy.

[0114] Specifically, the multi-dimensional analysis includes:

[0115] The processed school bus operation data is transferred to a high-dimensional feature space through mapping to extract key features of the data;

[0116] Set up analysis dimensions and create dimension prompts for each analysis dimension;

[0117] Use AI big models to conduct in-depth analysis of key data features and dimensional prompts, and summarize the analysis results of each dimension to obtain the current school bus operation analysis results.

[0118] like Figure 3 As shown, the artificial intelligence analysis unit is further described with a specific embodiment:

[0119] Design special dimension prompts based on different scenarios, such as departure and dispatch;

[0120] Generate training data using quantitative indicators and professional annotations, where the training data includes descriptions of actual situations and decision-making recommendations;

[0121] Map school bus operation data into a high-dimensional feature space and extract key features, such as vehicle spacing, number of passengers, and station requirements;

[0122] Set up analysis dimensions, such as departure decision dimension, vehicle scheduling dimension, passenger demand prediction dimension, and operation efficiency optimization dimension;

[0123] For each dimension, use the trained AI model to analyze, such as for the departure decision:

[0124] Input: {"distance to the preceding vehicle": 3,"number of vehicles in operation": 2,"average number of people in the vehicle": 10,"number of high-demand stops": 4,"average number of people on the vehicle": 5};

[0125] Output: Recommend departure, reason: many high-demand stations, the preceding vehicle is far away;

[0126] Based on the analysis results of various dimensions, a dispatching strategy is generated, such as {"departure decision":"depart immediately","route suggestion":"focus on station 3 and station 5","speed suggestion":"maintain normal speed"}.

[0127] In one embodiment of the present application, Lora fine-tuning is used when training the AI ​​big model to retain the language understanding ability of the AI ​​big model in the NLP field and enhance its decision-making ability in the vertical field of school bus scheduling planning. Determine the initial value of the hyperparameter for fine-tuning the AI ​​big model, and the default parameters are: number of training rounds = 2, number of gradient accumulation steps = 1, learning rate = 0.00001, batch = 8, cutoff length = 2048, and validation set ratio = 0.1.

[0128] A cloud application interface is configured to transmit the intelligent dispatching strategy and the processed school bus operation data to the mobile terminal, and the operator can query and use it in real time through the mobile terminal;

[0129] The intelligent scheduling execution unit is configured to transmit the intelligent scheduling strategy to the school bus driver, and the school bus driver allocates resources according to the intelligent scheduling strategy.

[0130] In one embodiment of the present application, when the school bus stops at each station during operation, the current data is synchronized and the intelligent scheduling strategy is returned to the school bus driver in the form of voice broadcast or message.

[0131] This application effectively suppresses interference and fluctuations during the charging process through a constant voltage filter cascade auto-disturbance rejection control strategy, and uses a first-level auto-disturbance rejection controller and a second-level auto-disturbance rejection controller to improve the robustness of wireless charging and improve charging efficiency. It also performs intelligent analysis of school bus operation data to achieve real-time dynamic scheduling, thereby improving school bus operation efficiency, optimizing resource allocation, and improving service quality.

[0132] like Figure 4 and Figure 5 As shown, the present invention provides an intelligent anti-interference control method for wireless charging school bus based on AI large model analysis, which is applied to the intelligent anti-interference wireless charging school bus control system as described above, and includes the following steps:

[0133] S1, taking the output voltage during charging as the control variable and setting the voltage as the input variable;

[0134] S2. Design the first-stage auto-disturbance rejection controller and the second-stage auto-disturbance rejection controller, and combine the first-stage Butterworth filter, the constant voltage filter cascade auto-disturbance rejection controller and the secondary side DC-DC circuit of the wireless charging unit into a constant voltage filter cascade auto-disturbance rejection control strategy;

[0135] S3, using constant voltage filter cascaded anti-disturbance control strategy to wirelessly charge the school bus;

[0136] S4, collect school bus operation data in real time, and transmit the school bus operation data to the central data processing platform for data processing;

[0137] S5. Perform multi-dimensional analysis on the processed school bus operation data based on the AI ​​big model and generate intelligent scheduling strategies;

[0138] S6, transmitting the intelligent dispatching strategy and the processed school bus operation data to the mobile terminal, and the operator conducts real-time query and use through the mobile terminal;

[0139] S7. Transmit the intelligent scheduling strategy to the school bus driver, and the school bus driver allocates resources according to the intelligent scheduling strategy.

[0140] It can be understood that the user layer collects the school bus operation data in the school bus card swiping system and sends it to the data center for data processing. The data center accepts the scheduled tasks sent by the scheduled task layer, and inputs the processed data into the AI ​​big model for training to obtain the trained AI big model and data analysis results. Users access and process data through the mini program, access the data analysis results of the data center, and determine the school bus usage based on the data analysis results.

[0141] like Figure 6 As shown, in one embodiment of the present application, four control methods are set, namely: first-order filtering LADRC control (F1-LADRC), second-order filtering LADRC control (F2-LADRC), first-order FCLADRC control (F1-CLADRC), and second-order FCLADRC control (F2-CLADRC):

[0142] The time taken for the startup process to reach the set value is 11.35ms, 10.24ms, 8.9ms, and 7.5ms for the four control strategies from high to low, namely F2-LADRC control, F1-LADRC control, F2-CLADRC control, and F1-CLADRC control. It can be seen that the constant voltage filter cascaded anti-disturbance control proposed in this application can effectively improve the response speed, but due to the increase in the filter order, the second-order controller is inferior to the first-order controller;

[0143] When the input voltage suddenly changes, the time it takes for the system to return to a steady state is shown in the following four control strategies: F2-LADRC control, F1-LADRC control, F2-CLADRC control, and F1-CLADRC control. It can be seen that the constant voltage filter cascaded anti-disturbance control applied can solve the problem of sudden input changes and reduce the time it takes for the system to return to a steady state. However, due to the increase in order, the system regulation becomes more complicated, resulting in a worse effect.

[0144] Faced with a sudden load change from 20 to 10. For the time it takes for the system to return to steady state, the four control strategies are F1-LADRC control, F2-LADRC control, F1-CLADRC control, and F2-CLADRC control from high to low. For the overshoot of the system, the four control strategies are F1-LADRC control, F2-LADRC control, F1-CLADRC control, and F2-CLADRC control from large to small. It can be seen that the constant voltage filter cascade self-anti-disturbance control of the present application can solve the problem of load mutation very well, among which the higher order has stronger anti-mutation ability than the lower order;

[0145] When the system faces an external step disturbance. For the time taken for the system to return to steady state, the four control strategies are F2-LADRC control, F1-LADRC control, F2-CLADRC control, and F1-CLADRC control from high to low. For the overshoot of the system, the four control strategies are F1-LADRC control, F2-LADRC control, F1-CLADRC control, and F2-CLADRC control from large to small. It can be seen that the constant voltage filter cascade self-anti-disturbance control of the present application can solve the problem of external disturbance very well, which verifies the effectiveness of the proposed strategy, in which the overshoot of a higher order is smaller than that of a lower order, and the time taken for a lower order to return to steady state is shorter than that of a higher order.

[0146] The present invention provides an electronic device, comprising: at least one processor, at least one memory, a communication interface and a bus; wherein the processor, memory and communication interface communicate with each other through the bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement the intelligent anti-interference wireless charging school bus control system as described above.

[0147] The present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to implement the intelligent anti-interference wireless charging school bus control system as described above.

[0148] 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 spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent anti-interference wireless charging school bus control system based on AI large model analysis, characterized in that: include: A wireless charging control unit, which is configured to include a primary circuit, a secondary circuit and a constant voltage filter cascaded auto-disturbance rejection control strategy, and uses the constant voltage filter cascaded auto-disturbance rejection control strategy to wirelessly charge the school bus, wherein the constant voltage filter cascaded auto-disturbance rejection control strategy includes a first-stage auto-disturbance rejection controller, a second-stage auto-disturbance rejection controller and a Butterworth filter; A data collection and transmission unit configured to collect school bus operation data in real time and transmit the school bus operation data to a central data processing platform for data processing; An artificial intelligence analysis unit, which is configured to perform multi-dimensional analysis on the processed school bus operation data based on the AI ​​big model and generate an intelligent dispatching strategy; A cloud application interface is configured to transmit the intelligent dispatching strategy and the processed school bus operation data to the mobile terminal, and the operator can query and use it in real time through the mobile terminal; An intelligent scheduling execution unit, which is configured to transmit the intelligent scheduling strategy to the school bus driver, and the school bus driver allocates resources according to the intelligent scheduling strategy; The constant voltage filter cascaded anti-disturbance control strategy specifically includes: The first-stage anti-disturbance controller processes the output voltage of the DC-DC circuit in the secondary circuit after passing through the Butterworth filter; the secondary circuit includes a rectifier output circuit and a DC-DC circuit; The second-stage ADRC processes the output voltage of the secondary-side DC-DC circuit; The Butterworth filter, the first-stage ADRC, the second-stage ADRC and the secondary-side DC-DC circuit are combined into a new composite cascade system to obtain a constant voltage filter cascade ADRC control strategy. The first-stage active disturbance rejection controller includes a first-stage linear extended state observer, a first state feedback controller, a first disturbance compensator and a first control rate generator, wherein: The two ends of the first-stage linear extended state observer are connected to the Butterworth filter and the first state feedback controller respectively, the other end of the first state feedback controller is connected to the first end of the first control rate generator, the second end of the first control rate generator is connected to the first disturbance compensator, and the first disturbance compensator is also connected to the second disturbance compensator. The calculation formula of the first-stage active disturbance rejection controller is: ； ； ； in, represents the change of state variables in the first-level linear extended state observer, represents the state matrix, , represents the input matrix, , represents the gain of the second-stage ADRC, represents the bandwidth parameter, represents the input of the Butterworth filter, represents the observed value of the first-stage ADRC, Indicates that Mapped to the output of the secondary circuit, represents the state vector of the first-level linear extended state observer, represents the acceleration output by the Butterworth filter, represents the total disturbance of the Butterworth filter, represents the control gain of the wireless charging unit, represents the output of the first control rate generator, For noise interference; The second-stage active disturbance rejection controller includes a second-stage linear extended state observer, a second state feedback controller, a second disturbance compensator, and a second control rate generator, wherein: The two ends of the second-stage linear extended state observer are respectively connected to the DC-DC circuit of the secondary circuit and the second state feedback controller, the other end of the second state feedback controller is connected to the first end of the second control rate generator, the second end of the second control rate generator is connected to the second disturbance compensator, the second disturbance compensator is connected to the secondary circuit through PWM, and the calculation formula of the second-stage active disturbance rejection controller is: ； ； in, represents the change of state variables in the second-level linear extended state observer, represents the state vector of the second-level linear extended state observer, represents the output noise of the secondary circuit, represents the observed value of the second-level ADRC, represents the output of the second control rate generator, Indicates that Mapped to the output of the secondary circuit, Represents the actual output of the second-stage ADRC; The calculation formula of the constant voltage filter cascade anti-disturbance control strategy is: ； ； ； ； ； ； in, express The rate of change, represents the state vector of the constant voltage filter cascaded active disturbance rejection control strategy, represents the input of the constant voltage filter cascaded auto-disturbance rejection control strategy, represents the actual output of the constant voltage filter cascaded anti-disturbance control strategy, represents the estimated output of the constant voltage filter cascaded active disturbance rejection control strategy, represents the proportional gain parameter of the first-stage auto-winding controller, express The state variable after the Butterworth filter, , and Respectively The three components of It represents the differential gain parameter of the first-stage auto-winding controller, represents the disturbance compensation gain parameter of the first-stage auto-winding controller, express The time derivative of represents the state matrix of the cascade system, where the cascade system includes a first-stage ADRC and a second-stage ADRC, represents the input matrix of the cascade system, represents the control input of the cascade system, represents the observer gain matrix of the cascade system, represents the estimated output of the cascade system, represents the actual output of the cascade system, represents the state vector of the cascade system, represents the control gain of the cascade system, represents the proportional gain parameter of the cascade system, represents the reference input, , and express The three components of represents the differential gain parameter of the cascade system, Represents the disturbance compensation gain parameter of the cascade system.

2. The intelligent anti-interference wireless charging school bus control system based on AI large model analysis as claimed in claim 1 is characterized in that: The school bus operation data includes card swiping data and vehicle data, wherein the card swiping data includes the card swiping time, card swiping location and card swiping type; the vehicle data includes the number of people in the car, the distance between the previous / rear school bus, the number of people in the previous / rear school bus, the recent average number of people getting on the bus at the alighting stop in different time periods, and the number of people getting on the bus at the previous school alighting stop.

3. The intelligent anti-interference wireless charging school bus control system based on AI large model analysis as described in claim 2 is characterized in that: The multi-dimensional analysis includes: The processed school bus operation data is transferred to a high-dimensional feature space through mapping to extract key features of the data; Set up analysis dimensions and create dimension prompts for each analysis dimension; Use AI big models to conduct in-depth analysis of key data features and dimensional prompts, and summarize the analysis results of each dimension to obtain the current school bus operation analysis results.

4. An intelligent anti-interference control method for wireless charging school bus based on AI large model analysis, characterized in that: The intelligent anti-interference wireless charging school bus control system as claimed in any one of claims 1 to 3 comprises the following steps: S1, taking the output voltage during charging as the control variable and setting the voltage as the input variable; S2. Design the first-stage auto-disturbance rejection controller and the second-stage auto-disturbance rejection controller, and combine the first-stage Butterworth filter, the constant voltage filter cascade auto-disturbance rejection controller and the secondary side DC-DC circuit of the wireless charging unit into a constant voltage filter cascade auto-disturbance rejection control strategy; S3, using constant voltage filter cascaded anti-disturbance control strategy to wirelessly charge the school bus; S4, collect school bus operation data in real time, and transmit the school bus operation data to the central data processing platform for data processing; S5. Perform multi-dimensional analysis on the processed school bus operation data based on the AI ​​big model and generate intelligent scheduling strategies; S6, transmitting the intelligent dispatching strategy and the processed school bus operation data to the mobile terminal, and the operator conducts real-time query and use through the mobile terminal; S7. Transmit the intelligent scheduling strategy to the school bus driver, and the school bus driver allocates resources according to the intelligent scheduling strategy.

5. An electronic device, characterized in that: include: At least one processor, at least one memory, a communication interface and a bus; wherein, The processor, memory, and communication interface communicate with each other via the bus; The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to implement the intelligent anti-interference wireless charging school bus control system as described in any one of claims 1-3.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to implement the intelligent anti-interference wireless charging school bus control system as described in any one of claims 1-3.

Images