Power supply modular combination method and system based on standardized interface

Abstract

The invention discloses a power supply modular combination method and system based on a standardized interface. The method comprises the following steps: step 1, collecting data through a sensor; 2, the central control hub fuses data and establishes a four-dimensional collaborative model; 3, the MOSFET driving circuit carries out adjustment according to a control instruction of the central control hub; and 4, the AI prediction current sharing module predicts the load change trend of the current power supply system and the performance degradation condition of each module through a built-in machine learning model, and dynamically adjusts the current distribution of each module based on the prediction result in combination with a target current value in an instruction, thereby realizing the current distribution of the power supply system. The system comprises a central control hub, a sensor unit, an AI prediction current sharing module and an MOSFET drive circuit. The method has the advantages that dynamic impedance matching and current sharing control closed loop are achieved, AI prediction current sharing is achieved through a machine learning model, output impedance and current distribution of a power module are accurately controlled, load changes are dynamically adapted, and current sharing precision and system stability are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power supply modules, in particular to a power supply modular combination method and system based on standardized interfaces. BACKGROUND

[0002] With the rapid development of new energy, industrial automation, data centers and other fields, higher requirements are put forward for the flexibility, reliability and efficiency of power supply systems. Traditional power supply systems have low modularization degree, complex fault maintenance and other problems, and are difficult to meet the diversified load demand and intelligent management demand. There are still technical bottlenecks in the multi-module collaborative control, dynamic performance optimization, predictive maintenance and other aspects of the power supply system. For example, the current sharing control precision between modules is insufficient, which leads to reduced system efficiency and shortened module life. The rapid reconstruction and adaptive adjustment ability is limited when a fault occurs, which affects the system reliability. In addition, with the deep integration of industrial internet and artificial intelligence technology, the power supply system urgently needs to have data-driven intelligent decision-making ability to realize multi-target dynamic balance and full life cycle management functions. SUMMARY

[0003] The purpose of the present application is to solve the problems of low modularization degree, complex fault maintenance, and difficulty in meeting the diversified load demand and intelligent management demand of traditional power supply systems. Therefore, a power supply modular combination method and system based on standardized interfaces are proposed.

[0004] The purpose of the present application can be achieved by the following technical solutions: a power supply modular combination method and system based on standardized interfaces, comprising:

[0005] Step 1: Collect output impedance data, current data, junction temperature data and fault signals through sensors, and transmit these data to the central control hub;

[0006] Step 2: The central control hub serves as a data fusion center, collects data feedback from sensors, builds a decision scheduling engine, establishes a four-dimensional collaborative model, iteratively optimizes the four-dimensional collaborative model through the decision scheduling engine, and converts the model optimization results into control instructions to feed back to the AI predictive current sharing module and MOSFET drive circuit;

[0007] Step 3: The MOSFET drive circuit receives the control instructions issued by the central control hub, adjusts the conduction degree and switching frequency of the MOSFET, changes the output impedance and current output of the power supply module, and feeds back the executed state data to the central control hub;

[0008] Step four: the AI prediction current sharing module predicts the load change trend and performance degradation of each module of the current power supply system through the built-in machine learning model, dynamically adjusts the current distribution of each module based on the prediction results and the target current value in the instruction, and feeds back the adjusted current distribution results to the central control hub for subsequent optimization and adjustment of the control strategy.

[0009] The application also provides a power supply modular combination system based on a standardized interface, and the above method is applied to the power supply modular combination system, which comprises a central control hub, a sensor unit, an AI prediction current sharing module and a MOSFET driving circuit.

[0010] Further, the sensor unit is used to collect data and transmit the data to the central control hub, and the sensor unit comprises a vector impedance sensor, a current sensor, a temperature sensor and a fault detection sensor, the vector impedance sensor is used to collect output impedance data of the power supply module, the current sensor is used to collect real-time current data, the temperature sensor is used to collect junction temperature data, and the fault detection sensor is used to collect fault signals.

[0011] Further, the central control hub serves as a data fusion center, collects the data transmitted by the sensor unit by means of clock synchronization technology, constructs a decision scheduling engine and establishes a four-dimensional collaborative model.

[0012] Further, the four-dimensional collaborative model is coupled with four key indicators, i.e., power supply output stability dimension, module life dimension, energy efficiency dimension and fault response dimension, to realize multi-objective dynamic balance, build an instruction distribution system by means of a CANFD bus and form a closed-loop control.

[0013] The established four-dimensional collaborative model can realize accurate control of the power supply output stability dimension, effectively reduce voltage fluctuation, provide more stable power supply for the load, meet the power demand of high-precision equipment, perform thermal management based on junction temperature data, monitor and adjust the module temperature in real time, avoid performance degradation and component damage caused by overheating, significantly prolong the service life of the power supply module, reduce maintenance costs, optimize real-time current and impedance, reduce the loss in the process of power transmission, improve the overall energy efficiency of the power supply system, meet the development trend of green energy saving, quickly process fault signals to effectively improve the fault tolerance and reliability of the system and ensure the continuity of power supply.

[0014] Further, the central control hub is built-in with a current sharing control algorithm for calculating the control parameters required for adjusting the output of the power supply module, thereby generating control instructions and issuing the control instructions to the MOSFET driving circuit to change the output impedance and current output of the power supply module by adjusting the MOSFET driving circuit.

[0015] By introducing a current error compensation term, real-time monitoring and correction of the current deviation between each power module is realized, so that the current distribution error between modules can be reduced to within ±1%, ensuring system load balancing and avoiding performance degradation or failure caused by partial module overload.

[0016] Further, the AI prediction current sharing module predicts the load change trend and performance degradation of each module through the built-in machine learning model, dynamically adjusts the current distribution of each module based on the predicted results and the target current value in the central control hub instruction, and feeds back the adjusted current distribution results to the central control hub for subsequent optimization and adjustment of the control strategy, thereby forming an inner-outer loop control.

[0017] By predicting the load change trend in advance, the current output is actively increased when the load is about to increase, which can effectively avoid voltage drop, and the current is reduced in time when the load decreases to prevent energy waste, greatly improving the dynamic response speed and stability of the power system. Dynamic current distribution according to the performance state of each module can intelligently compensate for performance-degrading modules, ensure system overall current sharing effect, reduce current imbalance problems caused by module performance differences, prolong the service life of power modules, and maintain system stable operation when some modules have performance degradation through dynamic current distribution strategy, reducing the impact of single-point failure on the overall system.

[0018] Further, the current sharing control algorithm is:

[0019]

[0020] where ΔD(k) is the conduction ratio adjustment amount of the kth iteration, μ(k) is the adaptive step size, is the output impedance gradient vector, λ is the current sharing weight coefficient, I avg (k) is the average output current of the module group, I i (k) is the current module output current.

[0021] Further, the way of dynamically adjusting the current distribution of each module is specifically:

[0022] When it is predicted that the load will increase, the current output of each module is increased in advance, and when the load decreases, the current of each module is reduced in proportion;

[0023] For modules with performance degradation, the current distribution proportion of the module is reduced according to its historical data and current state, and the reduced current is distributed to other modules to ensure the overall current sharing effect and stability of the system, and the adjusted current distribution results are fed back to the central control hub.

[0024] Further, the built-in machine learning model adopts a machine learning model that fuses a long short-term memory network and a support vector regression.

[0025] Compared with the prior art, the beneficial effects of the present application are:

[0026] 1. A four-dimensional coordination model based on an improved NSGA-II algorithm is constructed, and four dimensions of power output stability, module life, energy efficiency and fault response are coupled to realize multi-objective dynamic balance, improve the comprehensive performance of the power supply system through closed-loop control, prolong the service life of the module, reduce energy consumption, and quickly respond to faults to ensure stable operation of the system.

[0027] 2. Dynamic impedance matching and current sharing control closed loop, introducing current sharing error compensation term, using a fused machine learning model to realize AI prediction of current sharing, accurately controlling the output impedance and current distribution of the power module, dynamically adapting to load changes, compensating for module performance degradation, and improving current sharing accuracy and system stability.

[0028] 3. An AI-driven predictive maintenance system is built, a multi-modal deep neural network model containing life prediction, thermal runaway early warning and fault positioning sub-network is built, the module life, thermal runaway risk and fault location are predicted in advance, and the control strategy is linked to reduce the failure rate and improve system reliability. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to facilitate the understanding of those skilled in the art, the present application will be further described below with reference to the accompanying drawings.

[0030] Figure 1 The flowchart of the power supply modularization combination method based on the standardized interface of the present application.

[0031] The technical solutions of the present application will be described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application. Please refer to Figure 1 The power supply modularization combination method based on the standardized interface is as follows: Step 1: Collect output impedance data, current data, junction temperature data and fault signals through sensors, and transmit these data to the central control hub; Step 2: The central control hub serves as a data fusion center, collects the data fed back by the sensors, constructs a decision scheduling engine, establishes a four-dimensional coordination model, iteratively optimizes the four-dimensional coordination model through the decision scheduling engine, and converts the model optimization results into control instructions to feed back to the AI prediction current sharing module and the MOSFET drive circuit; Step three: the MOSFET drive circuit receives the control instruction issued by the central control hub, adjusts the conduction degree and switching frequency of the MOSFET, changes the output impedance and current output of the power module, and feeds back the executed state data to the central control hub; Step four: the AI prediction current sharing module predicts the load change trend and performance degradation of each module of the current power system through the built-in machine learning model, dynamically adjusts the current distribution of each module based on the prediction result and the target current value in the instruction, and feeds back the adjusted current distribution result to the central control hub for subsequent optimization and adjustment of the control strategy; The power module combination method of the application builds a data acquisition framework in step one, which provides data support for step two. Step two relies on step one to fuse the information provided by step one and establish a four-dimensional collaborative model for multi-objective dynamic balance. Step two feeds the generated control instruction back to step three based on the four-dimensional collaborative model. The MOSFET drive circuit in step three makes corresponding adjustments according to the control instruction. Step four predicts the load change trend and performance degradation of each module of the current power system, dynamically adjusts the current distribution of each module based on the prediction result and the target current value in the instruction of step two, thereby completing the process from data acquisition to instruction generation, instruction implementation, and prediction adjustment. The application also provides a power module combination system based on a standardized interface, and the above method is applied to the power module combination system. The system includes a central control hub, a sensor unit, an AI prediction current sharing module, and a MOSFET drive circuit. DETAILED DESCRIPTION

[0032] The technical solutions of the application will be described clearly and completely below with reference to the embodiments. Obviously, the described embodiments are only part of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0033] Please refer to Figure 1 The power module combination method based on a standardized interface is shown in the figure, and the steps are as follows:

[0034] Step one: collect output impedance data, current data, junction temperature data, and fault signals through sensors, and transmit these data to the central control hub;

[0035] Step two: the central control hub as a data fusion center, unified collection of sensor feedback data, build decision scheduling engine, establish four-dimensional collaborative model, through the decision scheduling engine on four-dimensional collaborative model iterative optimization, and the model optimization results into control instructions feedback to AI prediction current sharing module and MOSFET drive circuit;

[0036] Step three: MOSFET drive circuit receives the control instructions issued by the central control hub, by adjusting the on degree of MOSFET, switching frequency, change the output impedance and current output of power module, and the state data after execution feedback to the central control hub;

[0037] Step four: AI prediction current sharing module through the built-in machine learning model, the load change trend of current power supply system, the performance attenuation of each module is predicted, based on the prediction result, combined with the target current value in the instruction, dynamically adjust the current distribution of each module, at the same time, the current distribution result after adjustment is fed back to the central control hub, for the optimization adjustment of subsequent control strategy;

[0038] The power modular combination method builds a data acquisition framework through step one, step one provides data support for step two, step two relies on step one, and fuses the information provided by step one, and establishes a four-dimensional collaborative model for multi-target dynamic balance, step two feeds back the generated control instructions to step three based on the four-dimensional collaborative model, the MOSFET drive circuit in step three makes corresponding adjustment according to the control instructions, and step four predicts the load change trend of the current power supply system and the performance attenuation of each module, and dynamically adjusts the current distribution of each module based on the prediction result and the target current value in the instruction of step two, so as to complete the process from data acquisition to instruction generation, instruction implementation and prediction adjustment.

[0039] The application also provides a power modular combination system based on a standardized interface, and the above method is applied to the power modular combination system, and the system comprises a central control hub, a sensor unit, an AI prediction current sharing module and a MOSFET drive circuit.

[0040] The sensor unit comprises a vector impedance sensor, a current sensor, a temperature sensor and a fault detection sensor, and the sensor unit is used for collecting data and transmitting the data to the central control hub.

[0041] The central control hub serves as a data fusion center, acquires sensor unit signals by means of clock synchronization technology, constructs a decision scheduling engine based on an improved NSGA-II algorithm, and establishes a four-dimensional collaborative model.

[0042] The four-dimensional cooperative model is coupled by four key indicators of power supply output stability dimension, module life dimension, energy efficiency dimension and fault response dimension, realizes multi-objective dynamic balance, uses CANFD bus to build an instruction distribution system, and forms a closed-loop control;

[0043] The four-dimensional cooperative model construction process is: through clock synchronization technology to collect power module output impedance, real-time current, junction temperature data and fault signal raw data, power output voltage, filter denoising, normalization processing, eliminate dimension difference and abnormal value interference, ensure data quality;

[0044] The power output stability model: the voltage fluctuation range and ripple coefficient indicators are converted into quantitative parameters, the voltage fluctuation range is the difference between the maximum and minimum values of the actual measured voltage in a given time range, and the ripple coefficient is the ratio of the peak value of the AC ripple voltage to the DC output voltage. The voltage stability function is constructed, the least square method is used to fit the output voltage curve, and the standard deviation is calculated as the stability evaluation index;

[0045] Module life loss model: based on the junction temperature data, a thermal stress model is established, combined with the Arrhenius equation, the junction temperature and the module aging rate are related, and a life loss function is generated;

[0046] Energy efficiency optimization model: according to the real-time current and impedance data, a power loss model is constructed, the system energy loss is reduced by optimizing impedance matching, and an energy efficiency improvement objective function is established;

[0047] Fault suppression model: feature extraction is performed on the fault signal, the fault response time threshold and the processing priority are set, and the fault suppression function is constructed;

[0048] Using multi-objective optimization theory, the above four-dimensional quantitative function is used as a sub-objective, and a joint optimization model containing four sub-objects is constructed by using weighting coefficient method, and constraint conditions are introduced, including power supply output power range and module temperature safety interval, to ensure that the model meets the actual engineering requirements;

[0049] The improved NSGA-II algorithm is used to iteratively optimize the four-dimensional cooperative model, and the optimal solution set is searched in the feasible solution space through selection, crossover and mutation genetic operations, to realize the multi-objective dynamic balance of power supply output stability, module life, energy efficiency and fault response;

[0050] Using CANFD bus to build an instruction distribution system, the model optimization results are converted into control instructions and fed back to the power module, the real-time data after execution are collected for verification, and by comparing the target value with the actual output value, the model parameters and algorithm strategy are dynamically adjusted, forming a closed-loop control system of data acquisition, modeling optimization, instruction execution and effect feedback;

[0051] In use, the vector impedance sensor collects the output impedance data of the power module, the current sensor collects real-time current data, the temperature sensor collects temperature data, and the fault detection sensor collects fault signals, which are transmitted to the central control hub;

[0052] After receiving the data, the central control hub calculates the control parameters required to adjust the output of the power module based on the built-in four-dimensional collaborative model and the current sharing control algorithm, generates control instructions including current distribution instructions, and the current sharing control algorithm is ΔD(k) is the conduction ratio adjustment amount of the kth iteration, μ(k) is the adaptive step size, which is dynamically adjusted based on the fuzzy PID algorithm, is the output impedance gradient vector, λ is the current sharing weight coefficient, I avg (k) is the average output current of the module group, i (k) is the current module output current, and the instruction is based on the real-time current I i (k) of each module, avg (k) and the current sharing error compensation term λ·(I avg (k)-I i (k)) to determine the target current value that each module needs to output;

[0053] After receiving the current distribution instructions from the central control hub, the AI prediction current sharing module first predicts the load change trend and the performance degradation of each module through the built-in machine learning model, and based on the prediction results, adjusts the current distribution of each module dynamically in combination with the target current value in the instructions, including:

[0054] Load change response: when the load is predicted to increase, the current output of each module is increased by a preset value in advance to avoid voltage fluctuations due to insufficient current supply, and when the load decreases, the current of each module is reduced by a preset proportion to prevent energy waste;

[0055] Module performance compensation: for modules with performance degradation, the AI prediction current sharing module reduces the current distribution proportion of the module based on its historical data and current state, and distributes more current to other modules to ensure the overall current sharing effect and stability of the system, and at the same time, the adjusted current distribution result is fed back to the central control hub for subsequent optimization and adjustment of the control strategy, thereby forming an inner-outer loop control. MOSFET drive circuit receives the control instructions issued by the central control hub, adjusts the conduction degree and switching frequency of the MOSFET drive circuit, changes the output impedance and current output of the power module, executes the adjustment strategy of the impedance-current double closed loop controller, and feeds back the executed state data to the central control hub;

[0056] The module performance compensation process is specifically: the AI prediction current sharing module builds a module health degree evaluation model through historical data and real-time monitoring data, quantifies the module performance from three dimensions of output voltage stability, efficiency decay rate and temperature rise slope, adopts a weighted scoring method, gives the number of times of voltage fluctuation exceeding the threshold, the efficiency decline percentage and the junction temperature growth rate a weight of 0.4, 0.3 and 0.3 respectively, and calculates the module health degree score H i The lower the score, the more serious the performance degradation. According to the average value of all module health degree scores The modules lower than are defined as performance degradation modules, and the current distribution reduction ratio ΔP i of the performance degradation modules is inversely proportional to the health degree score, and the formula is α is an adjustment coefficient, the current reduced by the performance degradation module is redistributed according to the proportion of the health degree scores of other modules, that is, the current increment ΔI j of the performance normal module is While ensuring the overall current sharing effect of the system, the system efficiency and stability are maximized, and the adjusted current distribution result is fed back to the central control hub for optimization and adjustment of the subsequent control strategy, thereby forming an inner-outer loop control.

[0057] Where ΔI j represents the current increment of the jth power module with better performance, H j represents the health degree score of the jth power module, which reflects the current performance state of the module, and the higher the score, the better the module performance, ∑ k≠i H k is the sum of the health degree scores of all modules except the ith performance degradation module, which is used to determine the weight proportion of each module in current redistribution, ∑ i ΔP i represents the sum of the power reduction of all performance degradation modules, which represents the total amount of power lost due to module performance degradation, and I avg is the average current of the system, which is used to convert the power change into current adjustment amount.

[0058] The AI prediction current sharing module is also used for predictive maintenance processing, and the specific process is as follows:

[0059] Build a life prediction model: based on historical junction temperature data and current fluctuation, combined with Arrhenius equation and long short-term memory network, build a module life prediction model, the Arrhenius equation quantifies the influence of temperature on device aging rate, and the long short-term memory network processes time series data and learns the performance degradation trend of the module over time to predict the remaining service life.

[0060] Building a thermal runaway early warning model: using temperature gradient data collected in real time by temperature sensors, analyzing data features through convolutional neural networks, which are good at capturing spatial features, to identify abnormal temperature change patterns such as temperature mutations and overheating area diffusion, and to predict thermal runaway risks in advance;

[0061] Building a fault location model, based on fault signal features such as current mutations and voltage abnormal fluctuations, using graph neural networks to analyze the connection topology of power modules, and quickly locating the fault location through information transmission between nodes and edges to achieve accurate identification of faulty modules.

[0062] According to the prediction results, trigger the corresponding control mechanism, when the life prediction model judges that the remaining life of a module is lower than the threshold value, the central control hub adjusts the current distribution of the module and reduces its workload to the preset value, when the thermal runaway early warning model issues an alarm, the cooling system automatically increases the cooling power to the preset value, and the central control hub limits the output power of the power module, when the fault location model determines the fault module, the central control hub immediately starts the solid-state switch group, cuts off the fault loop and performs redundancy reconstruction.

[0063] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any skilled person in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application, therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0064] Finally: the above is only a preferred embodiment of the present application and is not intended to limit the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A modular power supply assembly method based on standardized interfaces, characterized in that, include: Step 1: Collect output impedance data, current data, junction temperature data, and fault signals through sensors, and transmit these data to the central control hub; Step 2: The central control hub, as the data fusion center, collects the data fed back by the sensors, builds a decision scheduling engine, establishes a four-dimensional collaborative model, iteratively optimizes the four-dimensional collaborative model through the decision scheduling engine, and converts the model optimization results into control commands to feed back to the AI ​​prediction current sharing module and MOSFET drive circuit. Step 3: The MOSFET driver circuit receives control commands from the central control hub, adjusts the MOSFET's conduction level and switching frequency to change the power module's output impedance and current output, and feeds back the executed status data to the central control hub. Step 4: The AI ​​predictive current sharing module uses a built-in machine learning model to predict the current load change trend of the power system and the performance degradation of each module. Based on the prediction results and the target current value in the command, it dynamically adjusts the current distribution of each module. At the same time, the adjusted current distribution results are fed back to the central control hub for subsequent optimization and adjustment of control strategies.

2. A modular power supply system based on a standardized interface, characterized in that: The system includes a central control hub, sensor units, an AI predictive current sharing module, and MOSFET drive circuitry.

3. The power supply modular combination system based on a standardized interface according to claim 2, characterized in that, The sensor unit is used to collect data and transmit the data to the central control hub. The sensor unit includes a vector impedance sensor, a current sensor, a temperature sensor, and a fault detection sensor. The vector impedance sensor is used to collect the output impedance data of the power module, the current sensor is used to collect real-time current data, the temperature sensor is used to collect junction temperature data, and the fault detection sensor is used to collect fault signals.

4. The power supply modular combination system based on a standardized interface according to claim 3, characterized in that, The central control hub, serving as a data fusion center, uses clock synchronization technology to collect data transmitted by sensor units, construct a decision scheduling engine, and establish a four-dimensional collaborative model.

5. The power supply modular combination system based on a standardized interface according to claim 4, characterized in that, The four-dimensional collaborative model achieves multi-objective dynamic balance by coupling and modeling four key indicators: power output stability, module lifespan, energy efficiency, and fault response. It also utilizes the CANFD bus to build an instruction distribution system, forming a closed-loop control.

6. The modular power supply system based on a standardized interface according to claim 2, characterized in that, The central control hub has a built-in current sharing control algorithm to calculate the control parameters required to adjust the power module output, thereby generating control commands to be sent to the MOSFET drive circuit. By adjusting the MOSFET drive circuit, the output impedance and current output of the power module are changed.

7. The power supply modular combination system based on a standardized interface according to claim 2, characterized in that, The AI ​​predictive current sharing module uses a built-in machine learning model to predict the load change trend of the power system and the performance degradation of each module. Based on the prediction results and the target current value in the central control hub command, it dynamically adjusts the current distribution of each module. At the same time, the adjusted current distribution results are fed back to the central control hub for subsequent optimization and adjustment of the control strategy, thus forming inner and outer loop control.

8. The power supply modular combination system based on a standardized interface according to claim 6, characterized in that, The flow sharing control algorithm is as follows: Where ΔD(k) is the conduction ratio adjustment in the k-th iteration, and μ(k) is the adaptive step size. Let I be the output impedance gradient vector, λ be the current sharing weighting coefficient, and I be the current sharing weighting coefficient. avg (k) represents the average output current of the module group, I i (k) represents the current output current of the module.

9. The power supply modular combination system based on a standardized interface according to claim 7, characterized in that, The specific method for dynamically adjusting the current distribution of each module is as follows: When an increase in load is anticipated, the current output of each module is increased in advance; when the load decreases, the current of each module is reduced proportionally. For modules whose performance has degraded, based on their historical data and current status, the current allocation ratio of the module is reduced, and the reduced current is allocated to other modules to ensure the overall current sharing effect and stability of the system. At the same time, the adjusted current allocation result is fed back to the central control hub.

10. The modular power supply system based on a standardized interface according to claim 7, characterized in that, The built-in machine learning model employs a fusion of long short-term memory network and support vector regression.

Images