A power management method and system for emergency support vehicles

Abstract

The present invention relates to the technical field of emergency power supply management. Specifically, the present invention relates to a power supply management method and system for an emergency support vehicle. The method collects exhaust gas flow rate, temperature and gas composition data through a ring sensor array to generate a heat flux density spatial map, uses a neural network prediction model to output a thermal attenuation warning map, starts a micro-resistance compensation unit in a graded manner according to the thermal energy gap value, and dynamically adjusts the circuit topology based on the heat flux density map. The output voltage is stabilized through a dual-channel feedback mechanism. The system includes a thermal field monitoring unit, a dynamic compensation unit and an intelligent voltage stabilization unit, which respectively realize thermal field data collection, graded thermal energy compensation and voltage stability control. This solution solves the problems of low thermal energy utilization efficiency and insufficient power supply stability in traditional technologies. Through dynamic analysis of the thermal field, intelligent compensation and adaptive voltage stabilization, the efficiency of temperature difference power generation is improved, and reliable power supply for emergency equipment is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency power supply management, and in particular to a power supply management method and system for an emergency support vehicle. Background Art

[0002] Emergency power management technology is an important technology. In scenarios such as emergency rescue and field operations, emergency support vehicle power management technology efficiently converts exhaust heat energy into electrical energy, and is the core means to ensure continuous power supply to critical loads such as communication equipment and medical equipment.

[0003] As the demand for emergency support develops towards high mobility and long endurance, building a power management system that integrates thermal field monitoring, intelligent compensation and voltage stabilization control is of key significance to improving energy utilization efficiency and ensuring the reliability of emergency power supply. The traditional extensive thermal energy utilization model has been difficult to adapt to the precise power supply needs under complex working conditions. However, the traditional emergency support vehicle power management technology has the core problems of low thermal energy utilization efficiency and insufficient power supply stability. The existing solution only relies on a single thermoelectric generator and does not establish a spatial distribution model of the exhaust thermal field. When the exhaust gas flow rate and temperature fluctuate, the difference in thermal energy attenuation in each area of ​​the hot end surface increases, resulting in a decrease in the efficiency of thermoelectric power generation. The use of a single solid The fixed resistance compensation method cannot dynamically match the real-time thermal energy gap. When the high-risk attenuation area appears, the compensation delay exceeds the preset range, which increases the effective thermal energy loss rate and reduces the battery heating efficiency. In addition, the traditional system lacks an adaptive voltage regulation mechanism, and the output voltage fluctuation amplitude increases, resulting in an increase in the malfunction rate of precision instruments and a significant increase in the risk of emergency equipment shutdown. This lack of thermal field analysis, dynamic compensation lag and insufficient voltage regulation capability ultimately make the traditional solution's comprehensive power supply efficiency low under complex working conditions, making it difficult to meet the high-reliability power supply requirements of emergency support scenarios. In order to solve this technical problem, we provide an emergency support vehicle power management method and system. Summary of the Invention

[0004] The purpose of the present invention is to provide a power management method and system for an emergency support vehicle to solve the problems raised in the above background technology.

[0005] Due to the low thermal energy utilization efficiency of traditional technologies, the thermal energy attenuation of different areas on the hot end surface varies greatly when the exhaust flow rate and temperature fluctuate, resulting in low thermoelectric power generation efficiency. Therefore, this case generates a spatial map of heat flux density, inputs a neural network prediction model to predict the thermal attenuation value, and activates a micro-resistance compensation unit to perform thermal energy compensation, which can improve the efficiency of thermoelectric power generation.

[0006] Because traditional systems suffer from insufficient power supply stability and large output voltage fluctuations, resulting in a high malfunction rate for precision instruments, this case establishes a dual-channel feedback regulation mechanism to adjust the circuit topology combination to stabilize the output voltage, thereby reducing the malfunction rate of precision instruments and ensuring the stability of power supply for emergency equipment.

[0007] To achieve the above objectives, one of the objectives of the present invention is to provide a method for managing power supply of an emergency support vehicle, comprising the following steps:

[0008] S1, using a ring-shaped sensor array to detect exhaust gas flow rate, temperature, and gas composition ratio in real time, and sending the detection data to the thermoelectric generator control module;

[0009] S2. Based on the data collected in S1, the thermal energy intensity value of each square centimeter of the hot end surface of the thermoelectric generator is calculated to generate a heat flux density spatial map. The heat flux density spatial map marks the thermal energy intensity value of each grid on the hot end surface of the thermoelectric generator. The heat flux density spatial map is input into a neural network prediction model. The neural network prediction model predicts the thermal attenuation value of each grid within the next five seconds by comparing historical thermal energy attenuation cases, outputs an early warning map marking high-risk attenuation areas, and calculates the effective thermal energy gap value of the high-risk attenuation area in the early warning map. According to the size of the effective thermal energy gap value, the micro-resistance compensation unit is activated in a graded manner to perform thermal energy compensation, and an electric energy output strategy is executed based on the heat flux density spatial map, that is, the output voltage is stabilized within the range of plus or minus 5% of the target voltage by adjusting the circuit topology combination;

[0010] S3. Supply the output voltage to the battery heating film, and heat the battery through the battery heating film.

[0011] As a further improvement of the present technical solution, the method for generating the heat flux density spatial map in S2 includes:

[0012] The hot end surface of the thermoelectric generator is divided into a 1cm×1cm grid array. The thermal energy intensity value of each grid is calculated according to the following logic:

[0013] The basic heat flow is obtained by multiplying the exhaust temperature value by the exhaust flow rate value, and the thermal conductivity is corrected according to the proportion of carbon dioxide in the gas composition ratio. The thermal energy intensity value = basic heat flow × thermal conductivity - heat loss caused by ambient temperature. The heat loss caused by ambient temperature is obtained by the sensor array, and the thermal energy intensity values ​​of each grid array are statistically analyzed to generate a heat flux density spatial map.

[0014] As a further improvement to this technical solution, the neural network prediction model performs the following operations:

[0015] The heat flux density spatial map and historical thermal energy decay cases are used as inputs to the neural network prediction model, and the spatial thermal decay features are extracted through three convolutional layers;

[0016] The first layer uses 16 3×3 convolution kernels to scan the thermal energy intensity values ​​of each grid array with a step size of 1 cm. Each convolution kernel calculates the weighted thermal gradient value of 9 grids in the coverage area and outputs the feature Figure 1 ;

[0017] The second layer features Figure 1 Apply 32 5×5 convolution kernels with a step size of 1 cm to calculate the thermal attenuation correlation factor of 25 units in each window and output the feature Figure 2 ;

[0018] The third layer features Figure 2 Eight 1×1 convolution kernels are used for feature compression to generate a thermal inertia coefficient matrix. Feature map 3 is output based on the thermal inertia coefficient matrix. The thermal inertia coefficient matrix of feature map 3 is input into the fully connected layer. The thermal attenuation value of each grid is output through the Sigmoid function. The thermal attenuation risk level is defined based on the thermal attenuation value, and a warning map is output to mark high-risk attenuation areas.

[0019] As a further improvement of this technical solution, the calculation method of the thermal energy gap value is:

[0020] Based on the coordinates of the high-risk attenuation area marked in the early warning map, the predicted thermal attenuation value of the corresponding grid is obtained, and the difference between the predicted thermal attenuation value and the preset safety threshold is calculated. When the difference result is a positive value, it is defined as the effective thermal energy gap value;

[0021] The logic of starting the micro-resistance compensation unit to perform thermal energy compensation according to the size of the effective thermal energy gap is as follows:

[0022] The corresponding level of compensation mode is triggered according to the preset range of the effective thermal energy gap value. When in the first range, the basic compensation mode is started to perform low-frequency pulse thermal energy supplementation. When in the second range, the medium-frequency continuous compensation mode is activated. When in the third range, the adjacent micro-resistance compensation units are synchronously called to form a collaborative compensation mode. The preset range is dynamically adjusted based on the analysis of historical working conditions data.

[0023] As a further improvement of the present technical solution, the method for performing thermal energy compensation by the micro-resistance compensation unit includes:

[0024] Based on the divided compensation mode and the grid corresponding to the high-risk attenuation area, the micro-resistance compensation unit is driven to move to the center position of the grid of the high-risk attenuation area. During the compensation process, the temperature change data of the high-risk attenuation area collected in real time by the sensor array is analyzed. When it is detected that the temperature rise rate of the high-risk attenuation area is lower than the preset standard temperature corresponding to the compensation mode, it is automatically upgraded to the next interval compensation mode, until the third interval;

[0025] In the third interval collaborative compensation mode, the micro-resistance compensation units in the high-risk attenuation area send synchronization instructions to the micro-resistance compensation units in the adjacent grid. The micro-resistance compensation units that receive the instructions adjust their output power so that the thermal energy compensation output by multiple micro-resistance compensation units is spatially superimposed, and the temperature balance of the superimposed area is verified through the heat flux density spatial map.

[0026] As a further improvement of this technical solution, the effect verification and feedback method of the thermal energy compensation is as follows:

[0027] After the thermal energy compensation cycle ends, the heat flux density distribution in the high-risk attenuation area is obtained through the sensor array, and the difference between the measured thermal energy intensity value and the expected value of thermal energy compensation is calculated. If the difference exceeds the preset tolerance range, the interval division standard and compensation mode trigger threshold of the next cycle are dynamically corrected according to the direction of the difference, and the corrected interval division standard and compensation mode trigger threshold of the next cycle are synchronized with the training data set of the neural network prediction model.

[0028] As a further improvement of this technical solution, the method for dynamically constructing the power output strategy includes:

[0029] According to the changing trend of the thermal energy intensity values ​​in each region of the heat flux density spatial map, the power generation units of the thermoelectric generator are divided into three categories: stable output type, fluctuation regulation type, and attenuation standby type. The circuit topology of the thermoelectric generator is reconstructed based on the classification results.

[0030] The stable output power generation units form a series main circuit, the fluctuation regulation power generation units are connected to the main circuit through parallel branches, and the attenuation standby power generation units are switched to the energy storage buffer mode. The classification standard is based on a comprehensive judgment of the variance of the thermal energy intensity value and the gradient change direction within the continuous monitoring period.

[0031] As a further improvement of the present technical solution, the method for stabilizing the output voltage within the range of plus or minus 5% of the target voltage by adjusting the circuit topology combination is as follows:

[0032] A dual-channel feedback regulation mechanism is established. The main channel monitors the deviation between the output voltage and the target voltage in real time based on the stable output power generation unit. When the absolute value of the deviation exceeds 5%, the number of power generation units in the series main circuit is adjusted by increasing or decreasing the number of power generation units, where the increase or decrease number is the absolute value of the deviation divided by the average output voltage value of a single power generation unit.

[0033] The secondary channel collects current fluctuation data from the sensor array through fast Fourier transform to extract high-frequency ripple components and low-frequency fluctuation components. A filtering circuit is connected to the parallel branch for the high-frequency component, and the phase compensation parameters between the parallel units are adjusted for the low-frequency component.

[0034] When the sensor array detects that the rate of change of the battery heating membrane current exceeds 5 amperes per second, the attenuated standby power generation unit is immediately switched from the energy storage buffer mode to the power supply state, and the voltage recovery stability is verified through the heat flux density spatial map.

[0035] As a further improvement of this technical solution, the output voltage stability optimization method includes:

[0036] Analyze the frequency domain characteristics of the output power quality, separate the high-frequency ripple and low-frequency fluctuation components, adopt a composite suppression strategy of electromagnetic shielding and filtering network for the high-frequency components, and design an adaptive phase compensation algorithm for the low-frequency components;

[0037] At the same time, a topology self-checking mechanism is established. When it is detected that the classification label of the power generation unit does not match the circuit connection status, the topology reconstruction is automatically triggered and the abnormal data is recorded for model optimization.

[0038] A second object of the present invention is to provide a system for implementing the above-mentioned emergency support vehicle power management method, comprising:

[0039] The thermal field monitoring unit consists of a ring-shaped sensor array, a thermoelectric generator hot-end grid array, and a heat flux density calculation module. It generates a real-time heat flux density spatial map and marks high-risk attenuation areas.

[0040] The dynamic compensation unit includes a micro-resistance compensation array, an electromagnetic drive positioning system, and a coordinated control module, which performs graded compensation and spatial thermal energy superposition according to the thermal energy gap value;

[0041] The intelligent voltage stabilization unit consists of a reconfigurable circuit topology, a dual-channel feedback controller, and an energy storage buffer circuit. It maintains output voltage stability through power generation unit classification and dynamic adjustment.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The present invention uses a ring sensor array to monitor the exhaust flow rate, temperature and gas composition in real time, generates a heat flux density spatial map and combines it with a neural network prediction model to accurately identify high-risk thermal attenuation areas, realize graded thermal energy compensation of the micro-resistance compensation unit, and improve the efficiency of thermoelectric power generation. At the same time, based on the heat flux density spatial map, the circuit topology of the thermoelectric power generation chip is dynamically classified and reconstructed, and the dual-channel feedback regulation mechanism is used to stabilize the output voltage within the target voltage range, effectively reducing the impact of voltage fluctuations on precision instruments, and improving the power utilization efficiency and power supply reliability of emergency support vehicles under complex working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is the overall workflow diagram of the present invention;

[0045] Figure 2It is a schematic diagram of the overall structure of the present invention.

[0046] The meaning of each number in the figure is:

[0047] 1. Thermal field monitoring unit; 2. Dynamic compensation unit; 3. Intelligent voltage stabilization unit. DETAILED DESCRIPTION

[0048] 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 the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0049] See also Figure 1 As shown, one of the purposes of this embodiment is to provide an emergency support vehicle power management method, including the following steps:

[0050] S1, using a ring-shaped sensor array to detect exhaust gas flow rate, temperature, and gas composition ratio in real time, and sending the detection data to the thermoelectric generator control module;

[0051] S2. Based on the data collected in S1, the thermal energy intensity value of each square centimeter of the hot end surface of the thermoelectric generator is calculated to generate a heat flux density spatial map. The heat flux density spatial map marks the thermal energy intensity value of each grid on the hot end surface of the thermoelectric generator. The heat flux density spatial map is input into a neural network prediction model. The neural network prediction model predicts the thermal attenuation value of each grid within the next five seconds by comparing historical thermal energy attenuation cases, outputs an early warning map marking high-risk attenuation areas, and calculates the effective thermal energy gap value of the high-risk attenuation area in the early warning map. According to the size of the effective thermal energy gap value, the micro-resistance compensation unit is activated in a graded manner to perform thermal energy compensation, and an electric energy output strategy is executed based on the heat flux density spatial map, that is, the output voltage is stabilized within the range of plus or minus 5% of the target voltage by adjusting the circuit topology combination;

[0052] In order to accurately quantify the heat energy distribution on the hot end surface of the thermoelectric generator for subsequent thermal attenuation prediction and compensation, a heat flux density spatial map needs to be generated. The method for generating the heat flux density spatial map in S2 includes:

[0053] Because the thermal energy intensity of different areas on the hot end surface of the thermoelectric power generation sheet varies spatially, the hot end surface of the thermoelectric power generation sheet is first divided into a 1 cm × 1 cm grid array. Taking a common 10 cm × 10 cm thermoelectric power generation sheet as an example, this division method can generate 100 grid cells, each of which corresponds to an independent thermal energy intensity calculation unit. The grid size is designed to balance calculation accuracy and efficiency. The 1 cm grid can capture subtle temperature fluctuations on the hot end surface while avoiding calculation redundancy caused by too small a grid. The thermal energy intensity value of each grid is calculated according to the following logic:

[0054] The basic heat flow is obtained by multiplying the exhaust gas temperature value by the exhaust gas flow rate value, wherein the exhaust gas temperature value is collected in real time by the temperature sensor in the annular sensor array, and the exhaust gas flow rate value is synchronously measured by the flow rate sensor. The basic heat flow reflects the basic heat energy carried by the exhaust gas flow, which is positively correlated with the actual heat conduction efficiency, and the thermal conductivity is corrected according to the proportion of carbon dioxide in the gas composition ratio. Since the proportion of carbon dioxide in the exhaust gas will affect the heat conduction efficiency, the thermal conductivity needs to be corrected according to the proportion of carbon dioxide in the gas composition ratio. Specifically, the preset thermal conductivity base value of pure exhaust gas is 1. When the carbon dioxide proportion increases by 1%, the thermal conductivity is increased by 0.02. The coefficient increases, then the thermal energy intensity value = basic heat flow × thermal conductivity - heat loss caused by ambient temperature. The heat loss caused by ambient temperature is obtained by the sensor array and quantified according to the temperature difference between the hot end surface of the thermoelectric generator and the environment. The thermal energy intensity value of each grid array is statistically analyzed and mapped into a two-dimensional spatial atlas according to the grid position. The color depth of each grid in the atlas is positively correlated with the thermal energy intensity value, and the specific value of each grid is marked to generate a heat flux density space atlas. The generation frequency of the atlas is 100 milliseconds / time, which can track the dynamic changes of the exhaust thermal field in real time and provide high-time resolution input data for the subsequent neural network prediction model.

[0055] In order to accurately predict the thermal attenuation value of each grid on the hot end surface of the thermoelectric generator and identify high-risk attenuation areas, a three-layer convolution feature extraction of the heat flux density spatial map is performed using a neural network prediction model. The neural network prediction model performs the following operations:

[0056] The grid thermal energy intensity value matrix of the heat flux density spatial map, that is, the 100-dimensional data of the 10×10 grid and the feature vector of the historical thermal energy attenuation case are dimensionally aligned and input into the model. The first convolution layer uses 16 3×3 convolution kernels to scan the grid array with a step size of 1 cm. Each convolution kernel covers 9 adjacent grids. The weighted thermal gradient value is obtained by calculating the weighted sum of the thermal energy intensity values ​​of each grid in the coverage area. The weight distribution follows the distance attenuation principle: the weight of the central grid is 0.4, and the weights of the adjacent peripheral grids decrease by 0.1 in turn, so as to highlight the dominant role of the thermal gradient change in the central area. All convolution kernels are calculated in parallel to generate 16 feature maps. Figure 1,Each feature map reflects the spatial distribution of a specific thermal gradient pattern;

[0057] The second layer features Figure 1 32 5×5 convolution kernels with a step size of 1 cm are used to calculate the thermal attenuation correlation factor of 25 units in each window. The thermal attenuation correlation factor is generated by calculating the correlation coefficient between the thermal gradient value of each grid in the window and the corresponding position in the historical attenuation case. The correlation calculation uses the Pearson correlation coefficient logic to measure the similarity between the current thermal gradient distribution and the historical attenuation pattern. The closer the coefficient value is to 1, the stronger the correlation. The 32 convolution kernels correspond to different historical attenuation types and output 32 feature maps. Figure 2 ,Each map characterizes the spatial correlation strength of a specific attenuation mechanism, enabling a preliminary classification of the potential causes of thermal attenuation;

[0058] The third layer features Figure 2 Eight 1×1 convolution kernels are used for dimensionality reduction processing. The high-dimensional features are compressed into a thermal inertia coefficient matrix through linear combination. The thermal inertia coefficient reflects the ability of each grid area to resist thermal attenuation. Its calculation is based on the feature Figure 2 The weighted integration of each correlation factor in the model assigns negative weights to features strongly associated with rapid decay modes and positive weights to features strongly associated with stable thermal fields, ultimately generating an 8-channel feature map 3. Each element of the thermal inertia coefficient matrix ranges from -1 to 1. A larger value indicates stronger thermal inertia and lower decay risk. The thermal inertia coefficient matrix of feature map 3 is expanded into a one-dimensional vector and input into a fully connected layer, which contains 128 neurons. The thermal inertia features are mapped to the thermal decay prediction space through the weight matrix. The output of the fully connected layer is output through the Sigmoid function to output the thermal decay value of each grid. The thermal decay risk level is defined based on the thermal decay value. The value is compressed to the range of 0 to 1 to obtain the thermal decay value of each grid. The closer the value is to 1, the higher the decay risk. The risk level is divided according to the thermal decay value: above 0.7 is high risk, 0.4 to 0.7 is medium risk, and below 0.4 is low risk. Finally, the risk level is mapped to the original grid coordinates to generate a warning map marking high-risk decay areas, providing intelligent decision support for the power supply thermal management of emergency support vehicles.

[0059] In order to accurately calculate the heat energy gap and implement graded compensation, the system locates the risk area based on the early warning map and dynamically adjusts the compensation strategy based on historical operating data. The heat energy gap value is calculated as follows:

[0060] First, the coordinate information of the corresponding grid is extracted from the high-risk attenuation area marked in the early warning map. Based on the coordinates of the high-risk attenuation area marked in the early warning map, the predicted thermal attenuation value of the corresponding grid is obtained from the thermal attenuation value matrix output by the neural network prediction model. This extraction process is implemented through matrix indexing operations to ensure the accurate correspondence between the coordinates and the thermal attenuation value. The predicted thermal attenuation value is calculated with the preset safety threshold. When the difference result is a positive value, it is defined as the effective thermal energy gap value, that is, effective thermal energy gap value = predicted thermal attenuation value - safety threshold. The effective thermal energy gap values ​​of all high-risk grids are arithmetic averaged to obtain the overall thermal energy gap value. This value reflects the severity of the current thermal attenuation risk and provides a quantitative basis for graded compensation.

[0061] The logic of starting the micro-resistance compensation unit to perform thermal energy compensation according to the size of the effective thermal energy gap is as follows:

[0062] Based on historical operating data, the effective thermal energy gap value is divided into three dynamic intervals. The first interval is when the thermal energy gap value is in the range of 0-0.2, corresponding to a mild thermal attenuation risk. The second interval is when the thermal energy gap value is in the range of 0.2-0.5, corresponding to a moderate thermal attenuation risk. The third interval is when the thermal energy gap value is in the range of 0.5-1.0, corresponding to a high thermal attenuation risk. The interval boundaries are updated every 24 hours based on the cluster analysis results of the latest 500 sets of operating data. The corresponding level of compensation mode is triggered according to the preset interval range of the effective thermal energy gap value. When in the first interval, the basic compensation mode is activated to perform low-frequency pulsed thermal energy supplement, that is, a single micro-resistance compensation unit is activated to perform low-frequency pulsed thermal energy supplement. The pulse frequency is set to 1 Hz, that is, a compensation pulse is turned on per second, each pulse lasts 200 milliseconds, and the output power is 30% of the rated power. This mode compensates for mild thermal attenuation through intermittent heating to avoid excessive energy compensation leading to excessive hot end temperature. When in the second interval, the medium-frequency continuous In compensation mode, the continuous output state of a single micro-resistance compensation unit is activated, and the output power is increased to 60% of the rated power. The compensation frequency is adjusted to 5 Hz, that is, it is turned on once every 200 milliseconds and lasts for 150 milliseconds. This mode quickly offsets moderate thermal attenuation through compensation with higher frequency and power. When in the third interval, the adjacent micro-resistance compensation units are synchronously called to form a collaborative compensation mode. The current micro-resistance compensation unit and its two adjacent micro-resistance compensation units, a total of 3 units, are synchronously called to form a collaborative compensation array. The output power of each unit is set to 80% of the rated power, and an alternating working mechanism is adopted: the first unit works for 300 milliseconds, the second one works for 300 milliseconds immediately, and the third one works for 400 milliseconds, forming a cyclic compensation with a period of 1 second. This mode copes with high thermal attenuation through multi-unit collaboration and higher power output. The preset interval range is dynamically adjusted based on historical operating condition data analysis, realizing intelligent response to thermal attenuation, forming a closed loop with the neural network prediction model, and effectively ensuring the stable operation of the emergency support vehicle power system.

[0063] To achieve accurate thermal compensation in high-risk attenuation areas, the micro-resistance compensation unit needs to be positioned according to the compensation mode and dynamically adjust the compensation strategy. The method for the micro-resistance compensation unit to perform thermal compensation includes:

[0064] Based on the divided compensation mode and the grid corresponding to the high-risk attenuation area, the micro-resistance compensation unit is driven to move to the center position of the high-risk attenuation area grid. According to the grid coordinates of the high-risk attenuation area marked in the early warning map, the geometric center coordinates of all grids in the area are calculated. The moving mechanism of the micro-resistance compensation unit is driven by a stepper motor to move to the center position with a positioning accuracy of 0.5 cm. The moving speed is set to 1 cm / s to ensure that the compensation unit is accurately aligned with the thermal attenuation area. After reaching the target position, the corresponding compensation mode is started according to the interval of the thermal energy gap value. If it is in the first interval, the basic compensation mode is started, the output power is 30% of the rated power, and the thermal energy is supplemented in a low-frequency pulse mode of 1 Hz. Each pulse lasts 200 milliseconds. If it is in the second interval, the medium-frequency continuous compensation mode is activated. The output power is increased to 60%, and the compensation frequency is adjusted to 5 Hz. It is turned on once every 200 milliseconds and lasts for 150 milliseconds. If it is in the third interval, the collaborative compensation mode is directly started, and the adjacent compensation units are called to form a collaborative array. During the compensation process, the temperature rise rate per unit time is calculated based on the temperature change data of the high-risk attenuation area collected in real time by the sensor array. Each compensation mode corresponds to a different preset standard temperature rise rate. Basic compensation mode: the preset standard is 1°C / second, medium-frequency continuous compensation mode: the preset standard is 2.5°C / second, and collaborative compensation mode: the preset standard is 4°C / second. When it is detected that the temperature rise rate in the high-risk attenuation area is lower than the preset standard temperature corresponding to the compensation mode, it is automatically upgraded to the next interval compensation mode until the third interval or the temperature recovery rate meets the requirements;

[0065] In the third interval collaborative compensation mode, the micro-resistance compensation unit in the high-risk attenuation area sends a synchronization instruction to the micro-resistance compensation unit in the adjacent grid. The instruction includes the compensation cycle and power output. The micro-resistance compensation unit that receives the instruction adjusts the output power according to the instruction of the main micro-resistance compensation unit, and adopts an alternating working mechanism: the main micro-resistance compensation unit works for 300 milliseconds first, the micro-resistance compensation unit on the right side works for 300 milliseconds, and the micro-resistance compensation unit on the lower side works for 400 milliseconds, forming a cycle of 1 second so that the thermal energy compensation output by multiple micro-resistance compensation units is superimposed in space, forming a superposition effect, and verifying the temperature balance of the superposition area through the heat flux density space map, and monitoring the temperature of the superposition area in real time through the heat flux density space map. The temperature balance index is calculated. The balance index is defined as the difference between the highest temperature and the lowest temperature in the superposition area. When the difference is less than 5°C, the temperature distribution is considered to be uniform and the compensation effect is good. If the difference exceeds 5°C, the power output ratio of each compensation unit is adjusted. For example, the power of the main micro-resistance compensation unit is reduced by 10%, and the power of the adjacent micro-resistance compensation unit is increased by 5% until the map shows that the temperature balance meets the standard, ensuring that the temperature in the thermal attenuation area recovers evenly. The thermal energy compensation method of the micro-resistance compensation unit achieves efficient compensation for high-risk thermal attenuation areas through precise positioning, dynamic mode upgrades and coordinated control, forming a closed loop with the heat flux density space map and neural network prediction model, effectively ensuring the thermal energy utilization efficiency of the emergency support vehicle power supply system.

[0066] To ensure the accuracy of thermal compensation and system adaptability, the thermal compensation effect verification and feedback method dynamically optimizes the compensation strategy and updates the model training data through difference analysis between measured data and expected values. The thermal compensation effect verification and feedback method is as follows:

[0067] After the thermal energy compensation cycle is completed, the sensor array is used to collect data from the high-risk attenuation area to obtain the heat flux density distribution in the high-risk attenuation area. The difference between the measured thermal energy intensity value and the expected thermal energy compensation value is calculated. If the difference exceeds the preset tolerance range, the interval division standard and compensation mode trigger threshold of the next cycle are dynamically corrected according to the difference direction. The expected thermal energy compensation value consists of two parts. The first part predicts the thermal attenuation value after compensation based on the heat flux density map and compensation mode before compensation and converts it into the expected thermal energy intensity value. The second part sets the ideal working temperature of each thermoelectric generator according to the ideal working temperature of the thermoelectric generator. The target thermal energy intensity value of the grid is taken as the larger value of the two as the final expected value to ensure that the compensation effect meets the power generation demand. The measured thermal energy intensity value of each grid is calculated point by point with the expected value of the corresponding grid. The absolute value of the difference of all high-risk grids is taken and the arithmetic mean is calculated to obtain the overall compensation error. The initial value of the preset tolerance range is 6% of the expected value, that is, when the absolute value of the difference exceeds 6% of the expected value, the correction is triggered. The value is dynamically adjusted by the standard deviation of the historical compensation data. When the overall compensation error exceeds the tolerance range, the measured value is adjusted according to the direction of the difference (the measured value> the expected value is a positive deviation, otherwise it is a negative deviation). ) initiates corrections. Positive deviations (overcompensation) lower the compensation power thresholds for each interval. Negative deviations (undercompensation) increase the sensitivity of the compensation mode. A clustering algorithm is used to regroup historical compensation data, dynamically adjusting the boundaries of the three intervals. Effective thermal energy gap values ​​from the past 50 compensation cycles are collected, and K-means clustering is used to divide the data into three categories, corresponding to the three intervals. Based on the cluster center, the new interval boundaries are expanded by 15% on both sides to make the interval division more consistent with the compensation requirements of the current operating conditions. The revised interval division criteria for the next cycle and the compensation mode trigger threshold are synchronized with the training dataset of the neural network prediction model. The interval division criteria serve as feature scaling parameters for the model input, and the trigger threshold serves as a constraint condition for the model output, such as the mapping relationship between the compensation mode and the output thermal attenuation value. The training dataset of the neural network prediction model is updated using incremental learning, retaining the original 80% of historical valid data and adding the latest 20% of corrected parameters and corresponding compensation cases. The fully connected layer parameters of the model are retrained, while the convolutional layer parameters remain unchanged to avoid destroying the learned spatial features, effectively ensuring the stable and efficient operation of the emergency support vehicle's temperature difference power generation system.

[0068] To achieve efficient power output from thermoelectric generators, a dynamic power output strategy based on the heat flux density spatial map optimizes the conversion of thermal energy into electrical energy by classifying power generation units and reconstructing circuit topology. The dynamic construction method of the power output strategy includes:

[0069] The thermal energy intensity values ​​of the grids corresponding to each power generation unit are continuously collected, and the variance and gradient change direction are calculated. The variance calculation is used to measure the fluctuation amplitude of the thermal energy intensity. The formula is the average of the sum of the squares of the differences between each data point and the mean. The smaller the variance, the more stable the thermal energy. The gradient direction calculation is determined by the sign of the thermal energy intensity difference between adjacent cycles (current value - previous cycle value, where the result greater than 0 indicates an increase, less than zero indicates a decrease, and equal to zero indicates stability). According to the change trend of the thermal energy intensity values ​​of each area in the heat flux density space map, the power generation units of the thermoelectric power generation sheet are divided into three categories: stable output type, fluctuation regulation type and attenuation standby type. The specific classification standards for the three types of power generation units are: stable output type is thermal energy The variance of the intensity value is less than 0.15, and the gradient change direction is stable for 10 consecutive cycles, accounting for about 30%-40%, corresponding to the area with uniform color and no obvious change in the heat flux density map. The variance of the fluctuation regulation type power generation unit is between 0.15-0.4, and the gradient change direction shows periodic fluctuations, such as alternating rise and fall, corresponding to the area with alternating light and dark colors in the map, which is mostly caused by exhaust gas flow rate fluctuations. The variance of the attenuation standby type power generation unit is greater than 0.4, and the gradient change direction is continuously declining for 5 consecutive cycles, corresponding to the high-risk attenuation area with significantly lighter colors in the map. Thermal energy can no longer support effective power generation, and the circuit topology of the thermoelectric generator is reconstructed based on the classification results.

[0070] The stable output power generation units form a series main circuit. The stable output power generation units are sorted from high to low according to the heat flux density and connected in series through solid-state relays to form a main circuit. The series connection can accumulate the output voltage of each unit. The series connection order is arranged in descending order of thermal energy intensity, so that the high thermal energy unit has priority in the power generation task, thereby improving the overall output efficiency of the main circuit. The fluctuation regulation power generation unit is connected to the main circuit through a parallel branch. The fluctuation regulation unit forms a parallel branch through a bidirectional thyristor. Each branch contains 2 to 3 units and is connected to the main circuit through a DC-DC converter. When the thermal energy intensity of a branch increases, the thyristor is turned on to merge the branch into the main circuit to supplement the output power. When the thermal energy intensity decreases, the thyristor is turned off to cut off the branch to avoid reverse current loss. The decaying standby power generation unit switches to the energy storage buffer mode. The decaying standby unit switches to the energy storage buffer circuit through the MOSFET switch and connects to the supercapacitor. When the unit's thermal energy intensity decays, the MOSFET turns on and stores the residual energy in the supercapacitor. When the output of other units fluctuates, the supercapacitor releases energy to compensate for the power gap. The classification standard is based on the comprehensive judgment of the variance and gradient change direction of the thermal energy intensity value within the continuous monitoring period. The power generation unit classification is started every 10 seconds. When one of the following conditions is met, reclassification is forced to be triggered: the main circuit output voltage fluctuation exceeds ±15%, the proportion of decaying standby units exceeds 50% of the total number of units, and the area of ​​the high-risk area in the heat flux density map is expanded by more than 2 times. The real-time adjustment process of the circuit topology is as follows:

[0071] The classification algorithm generates a unit classification list with a type label for each unit. The control chip generates a switch control signal based on the list. The solid-state relay of the stable output power generation unit is closed to form a series circuit. The bidirectional thyristor of the fluctuation regulation unit is dynamically turned on / off according to the real-time thermal energy intensity. The MOSFET of the attenuation standby unit is switched to energy storage mode. After adjustment, the output parameters are monitored in real time through voltage and current sensors. If the voltage deviation exceeds ±5%, the series sequence or the number of parallel branches is re-optimized to significantly enhance its adaptability under fluctuating heat flux conditions, providing an efficient and reliable power output solution for the power supply system of the emergency support vehicle.

[0072] To stabilize the output voltage of the thermoelectric generator within the range of ±5% of the target voltage, the system dynamically optimizes the circuit topology through a dual-channel feedback regulation mechanism and an emergency switching strategy. The method for stabilizing the output voltage within the range of ±5% of the target voltage by adjusting the circuit topology combination is as follows:

[0073] A dual-channel feedback regulation mechanism was established. The main channel, based on stable output power generation units, monitors the deviation between the output voltage and the target voltage in real time. When the absolute value of the deviation exceeds 5%, adjustments are made by increasing or decreasing the number of power generation units in the series main circuit. The increase or decrease is calculated by dividing the absolute value of the deviation by the average output voltage of a single power generation unit. Solid-state relays are then used to disconnect the corresponding number of units, reducing the total output voltage. After adjustment, monitoring is repeated until the deviation returns to within ±5%.

[0074] The secondary channel collects current fluctuation data from the sensor array through fast Fourier transform to extract high-frequency ripple components and low-frequency fluctuation components. A filtering circuit is connected to the parallel branch for the high-frequency component, and the phase compensation parameters between the parallel units are adjusted for the low-frequency component.

[0075] When the sensor array detects that the current change rate of the battery heating membrane exceeds 5 amperes per second, the attenuation standby power generation unit is immediately switched from the energy storage buffer mode to the power supply state. The switching logic is to disconnect the energy storage supercapacitor connection, connect the attenuation unit to the main circuit through the bypass diode, and adjust its output power to 30% of the rated value to avoid overload. The switching process is completed within 10 milliseconds and can provide an additional 0.5-1V voltage compensation. After the switch, the heat flux density space map is updated in real time to monitor the temperature recovery and voltage stability of the corresponding area of ​​the attenuation unit and verify the voltage recovery stability through the heat flux density space map. The voltage recovery stability verification standard is: the voltage fluctuation amplitude is less than ±2% within 1 second after the switch, the temperature change rate of the attenuation unit area in the heat flux density map is less than 0.5℃ / second within 3 seconds, and the supercapacitor energy storage capacity is restored to more than 80% within 5 seconds after the switch. If the conditions are not met, one attenuation unit is automatically added until the voltage stabilizes to ensure the reliability of the emergency power supply.

[0076] To optimize the stability of the thermoelectric generator's output voltage, the system implements multi-dimensional optimization through frequency domain feature analysis, composite suppression strategy, adaptive phase compensation, and topology self-checking mechanism. The output voltage stability optimization methods include:

[0077] The output voltage signal is collected in real time through a high-precision voltage sensor, and the time domain signal is converted into the frequency domain using the fast Fourier transform to obtain the voltage spectrum distribution. The frequency domain characteristics of the output power quality are analyzed based on the fast Fourier transform results, and the high-frequency ripple and low-frequency fluctuation components are separated. A composite suppression strategy of electromagnetic shielding and filtering network is adopted for the high-frequency component, and an adaptive phase compensation algorithm is designed for the low-frequency component. At the same time, a topology structure self-check mechanism is established. When it is detected that the classification label of the power generation unit does not match the circuit connection status, the topology reconstruction is automatically triggered and the abnormal data is recorded for model optimization, which effectively improves the power supply quality and reliability of the emergency support vehicle power system.

[0078] S3. Supply the output voltage to the battery heating film, and heat the battery through the battery heating film.

[0079] A second object of the present invention is to provide a system for implementing a power management method for an emergency support vehicle, comprising:

[0080] The thermal field monitoring unit 1 consists of a ring-shaped sensor array, a thermoelectric generator hot end grid array, and a heat flux density calculation module. It generates a heat flux density spatial map in real time and marks high-risk attenuation areas.

[0081] The dynamic compensation unit 2 includes a micro-resistance compensation array, an electromagnetic drive positioning system, and a coordinated control module, which performs graded compensation and spatial thermal energy superposition according to the thermal energy gap value;

[0082] The intelligent voltage stabilization unit 3 is composed of a reconfigurable circuit topology, a dual-channel feedback controller and an energy storage buffer circuit, and maintains output voltage stability through power generation unit classification and dynamic adjustment.

[0083] This method collects exhaust flow rate, temperature and gas composition data through a ring sensor array to generate a heat flux density spatial map, uses a neural network prediction model to output a thermal attenuation warning map, starts the micro-resistance compensation unit in stages according to the thermal energy gap value, and dynamically adjusts the circuit topology based on the heat flux density map. The output voltage is stabilized through a dual-channel feedback mechanism. The system includes a thermal field monitoring unit, a dynamic compensation unit and an intelligent voltage stabilization unit, which respectively realize thermal field data collection, hierarchical thermal energy compensation and voltage stability control. This solution solves the problems of low thermal energy utilization efficiency and insufficient power supply stability in traditional technologies. Through dynamic analysis of the thermal field, intelligent compensation and adaptive voltage stabilization, the efficiency of temperature difference power generation is improved, and the reliable power supply of emergency equipment is guaranteed.

[0084] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A power management method for an emergency support vehicle, characterized by: The following steps are involved: S1, using a ring-shaped sensor array to detect exhaust gas flow rate, temperature, and gas composition ratio in real time, and sending the detection data to the thermoelectric generator control module; S2. Based on the data collected in S1, the thermal energy intensity value of each square centimeter of the hot end surface of the thermoelectric generator is calculated to generate a heat flux density spatial map. The heat flux density spatial map marks the thermal energy intensity value of each grid on the hot end surface of the thermoelectric generator. The heat flux density spatial map is input into a neural network prediction model. The neural network prediction model predicts the thermal attenuation value of each grid within the next five seconds by comparing historical thermal energy attenuation cases, outputs an early warning map marking high-risk attenuation areas, and calculates the effective thermal energy gap value of the high-risk attenuation area in the early warning map. According to the size of the effective thermal energy gap value, the micro-resistance compensation unit is activated in a graded manner to perform thermal energy compensation, and an electric energy output strategy is executed based on the heat flux density spatial map, that is, the output voltage is stabilized within the range of plus or minus 5% of the target voltage by adjusting the circuit topology combination; S3. Supply the output voltage to the battery heating film, and heat the battery through the battery heating film.

2. The method for managing power supply of an emergency support vehicle according to claim 1, characterized in that: The method for generating the heat flux density spatial map in S2 includes: The hot end surface of the thermoelectric generator is divided into a 1cm×1cm grid array. The thermal energy intensity value of each grid is calculated according to the following logic: The basic heat flow is obtained by multiplying the exhaust temperature value by the exhaust flow rate value, and the thermal conductivity is corrected according to the proportion of carbon dioxide in the gas composition ratio. The thermal energy intensity value = basic heat flow × thermal conductivity - heat loss caused by ambient temperature. The heat loss caused by ambient temperature is obtained by the sensor array, and the thermal energy intensity values ​​of each grid array are statistically analyzed to generate a heat flux density spatial map.

3. The method for managing power supply of an emergency support vehicle according to claim 2, characterized in that: The neural network prediction model performs the following operations: The heat flux density spatial map and historical thermal energy decay cases are used as inputs to the neural network prediction model, and the spatial thermal decay features are extracted through three convolutional layers; The first layer uses 16 3×3 convolution kernels to scan the thermal energy intensity values ​​of each grid array with a step size of 1 cm. Each convolution kernel calculates the weighted thermal gradient value of 9 grids in the coverage area and outputs feature map 1; The second layer applies 32 5×5 convolution kernels to feature map 1 with a step size of 1 cm to calculate the thermal attenuation correlation factor of 25 units in each window and output feature map 2; The third layer uses 8 1×1 convolution kernels to compress the features of feature image 2 to generate a thermal inertia coefficient matrix. The thermal inertia coefficient matrix of feature image 3 is output based on the thermal inertia coefficient matrix. The thermal inertia coefficient matrix of feature image 3 is input into the fully connected layer, and the thermal attenuation value of each grid is output through the Sigmoid function. The thermal attenuation risk level is defined based on the thermal attenuation value, and a warning map is output to mark high-risk attenuation areas.

4. The method for managing power supply of an emergency support vehicle according to claim 3, characterized in that: The calculation method of the thermal energy gap value is: Based on the coordinates of the high-risk attenuation area marked in the early warning map, the predicted thermal attenuation value of the corresponding grid is obtained, and the difference between the predicted thermal attenuation value and the preset safety threshold is calculated. When the difference result is a positive value, it is defined as the effective thermal energy gap value; The logic of starting the micro-resistance compensation unit to perform thermal energy compensation according to the size of the effective thermal energy gap is as follows: The corresponding level of compensation mode is triggered according to the preset range of the effective thermal energy gap value. When in the first range, the basic compensation mode is started to perform low-frequency pulse thermal energy supplementation. When in the second range, the medium-frequency continuous compensation mode is activated. When in the third range, the adjacent micro-resistance compensation units are synchronously called to form a collaborative compensation mode. The preset range is dynamically adjusted based on the analysis of historical working conditions data.

5. The method for managing power supply of an emergency support vehicle according to claim 4, characterized in that: The method for performing thermal energy compensation by the micro-resistance compensation unit includes: Based on the divided compensation mode and the grid corresponding to the high-risk attenuation area, the micro-resistance compensation unit is driven to move to the center position of the grid of the high-risk attenuation area. During the compensation process, the temperature change data of the high-risk attenuation area collected in real time by the sensor array is analyzed. When it is detected that the temperature rise rate of the high-risk attenuation area is lower than the preset standard temperature corresponding to the compensation mode, it is automatically upgraded to the next interval compensation mode, until the third interval; In the third interval collaborative compensation mode, the micro-resistance compensation units in the high-risk attenuation area send synchronization instructions to the micro-resistance compensation units in the adjacent grid. The micro-resistance compensation units that receive the instructions adjust their output power so that the thermal energy compensation output by multiple micro-resistance compensation units is spatially superimposed, and the temperature balance of the superimposed area is verified through the heat flux density spatial map.

6. The method for managing power supply of an emergency support vehicle according to claim 5, characterized in that: The effect verification and feedback method of the thermal energy compensation is as follows: After the thermal energy compensation cycle ends, the heat flux density distribution in the high-risk attenuation area is obtained through the sensor array, and the difference between the measured thermal energy intensity value and the expected value of thermal energy compensation is calculated. If the difference exceeds the preset tolerance range, the interval division standard and compensation mode trigger threshold of the next cycle are dynamically corrected according to the direction of the difference, and the corrected interval division standard and compensation mode trigger threshold of the next cycle are synchronized with the training data set of the neural network prediction model.

7. The method for managing power supply of an emergency support vehicle according to claim 6, characterized in that: The dynamic construction method of the power output strategy includes: According to the changing trend of the thermal energy intensity values ​​in each region of the heat flux density spatial map, the power generation units of the thermoelectric generator are divided into three categories: stable output type, fluctuation regulation type, and attenuation standby type. The circuit topology of the thermoelectric generator is reconstructed based on the classification results. The stable output power generation units form a series main circuit, the fluctuation regulation power generation units are connected to the main circuit through parallel branches, and the attenuation standby power generation units are switched to the energy storage buffer mode. The classification standard is based on a comprehensive judgment of the variance of the thermal energy intensity value and the gradient change direction within the continuous monitoring period.

8. The method for managing power supply of an emergency support vehicle according to claim 7, characterized in that: The method for stabilizing the output voltage within the range of plus or minus 5% of the target voltage by adjusting the circuit topology combination is as follows: A dual-channel feedback regulation mechanism is established. The main channel monitors the deviation between the output voltage and the target voltage in real time based on the stable output power generation unit. When the absolute value of the deviation exceeds 5%, the number of power generation units in the series main circuit is adjusted by increasing or decreasing the number of power generation units, where the increase or decrease number is the absolute value of the deviation divided by the average output voltage value of a single power generation unit. The secondary channel collects current fluctuation data from the sensor array through fast Fourier transform to extract high-frequency ripple components and low-frequency fluctuation components. A filtering circuit is connected to the parallel branch for the high-frequency component, and the phase compensation parameters between the parallel units are adjusted for the low-frequency component. When the sensor array detects that the rate of change of the battery heating membrane current exceeds 5 amperes per second, the attenuated standby power generation unit is immediately switched from the energy storage buffer mode to the power supply state, and the voltage recovery stability is verified through the heat flux density spatial map.

9. The method for managing power supply of an emergency support vehicle according to claim 8, characterized in that: Methods for optimizing output voltage stability include: Analyze the frequency domain characteristics of the output power quality, separate the high-frequency ripple and low-frequency fluctuation components, adopt a composite suppression strategy of electromagnetic shielding and filtering network for the high-frequency components, and design an adaptive phase compensation algorithm for the low-frequency components; At the same time, a topology self-checking mechanism is established. When it is detected that the classification label of the power generation unit does not match the circuit connection status, the topology reconstruction is automatically triggered and the abnormal data is recorded for model optimization.

10. A system for implementing the emergency support vehicle power management method according to any one of claims 1 to 9, characterized in that: include: The thermal field monitoring unit (1) consists of a ring-shaped sensor array, a thermoelectric generator hot end grid array, and a heat flux density calculation module, which generates a heat flux density spatial map in real time and marks high-risk attenuation areas; The dynamic compensation unit (2) includes a micro-resistance compensation array, an electromagnetic drive positioning system and a coordinated control module, and performs hierarchical compensation and spatial thermal energy superposition according to the thermal energy gap value; The intelligent voltage stabilization unit (3) is composed of a reconfigurable circuit topology, a dual-channel feedback controller and an energy storage buffer circuit, and maintains output voltage stability through power generation unit classification and dynamic adjustment.

Images