Shelter intelligent power supply control method and system based on 5G communication

Abstract

The invention provides a distributed energy optimization management method and system for rapidly deploying a mobile shelter hospital. The method comprises the following steps: receiving the working state, power demand and external environment data of electric equipment in a shelter in real time through a 5G network; and integrating the information by using a multi-source data fusion technology to obtain a comprehensive data set. A deep learning algorithm is adopted to analyze a comprehensive data set, power demand prediction is generated, real-time electricity price information is combined, a genetic algorithm is used to optimize equipment starting time and power output, a control plan is made, a control instruction is sent to the intelligent power distribution module through a 5G network, and accurate power supply management is achieved. And simulating the operation data by using a reinforcement learning technology to generate a control strategy. The state of a power system is monitored through a 5G network, actual operation data are collected, a control scheme is adjusted in combination with a new strategy, and an intelligent power supply control scheme is formed. According to the invention, the power management efficiency and intelligent level of shelter facilities are improved, and the reliability and economical efficiency of the system are enhanced.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of communication technology, and in particular to a method and system for controlling an intelligent power supply of a cabin based on 5G communication. Background Art

[0002] With the rapid development of mobile medical care, emergency rescue and other fields, the application of portable facilities such as shelter hospitals and temporary command centers is becoming more and more widespread. These facilities require a highly reliable power supply system to support the normal operation of various key equipment during operation. Especially in remote areas or emergency situations, how to efficiently use limited power resources has become an important issue. Therefore, it is particularly important to develop a system that can monitor and intelligently manage power usage in shelters in real time.

[0003] At present, there are some power management systems based on traditional communication technology and simple algorithms on the market, which mainly rely on preset rules to control the switch and adjust the power of the equipment. Such systems usually include basic data acquisition modules, processing units and actuators, which can achieve effective allocation of power resources to a certain extent. In addition, some advanced systems have also tried to introduce preliminary artificial intelligence elements, such as using simple machine learning models to predict short-term power consumption trends and adjust power supply plans accordingly; most of the existing solutions can only process data from a single source, lacking the ability to conduct comprehensive analysis of multi-source heterogeneous information, such as equipment status, external environmental conditions, etc., resulting in incomplete and inaccurate decision-making; most of the current systems use rule-based methods or relatively simple prediction models, which are not efficient in solving the optimal solution in complex situations and are difficult to meet the needs of refined management. Summary of the invention

[0004] The embodiment of the present invention provides a method and system for intelligent power supply control of a cabin based on 5G communication, which is used to solve the problems that most existing solutions in the prior art can only process data from a single source, lack the ability to conduct comprehensive analysis on multi-source heterogeneous information, resulting in incomplete and inaccurate decision-making, and have low efficiency in finding the optimal solution in complex situations, making it difficult to meet the needs of refined management.

[0005] In a first aspect, an embodiment of the present invention provides a method for controlling a cabin intelligent power supply based on 5G communication, comprising:

[0006] Receive the working status information and power demand signals of each electrical device in the shelter and the external environment data of the shelter in real time through the 5G network;

[0007] The working status information, power demand signal and external environment data of the shelter are integrated by using multi-source data fusion technology to obtain a comprehensive data set;

[0008] A deep learning algorithm is used to analyze the comprehensive data set to generate a power demand forecast result, and the power demand forecast result is combined with real-time power market electricity price information, and a genetic algorithm is used to optimize the start-up time and power output of different electrical equipment in the cabin to form an equipment control plan;

[0009] Based on the equipment control plan, control instructions are output to the intelligent power distribution module in the cabin through the 5G network to achieve accurate power supply management for each electrical equipment;

[0010] Using reinforcement learning technology to simulate the operating data in the precise power supply management process to generate a new control strategy;

[0011] The operating status of the power system in the cabin is continuously monitored through the 5G network, actual operating data is collected, the actual operating data is combined with the new control strategy, the control strategy is adjusted, and an intelligent power supply control solution is formed.

[0012] Optionally, a deep learning algorithm is used to analyze the comprehensive data set to generate a power demand forecast result, and the power demand forecast result is combined with real-time power market electricity price information, and a genetic algorithm is used to optimize the start-up time and power output of different electrical equipment in the shelter to form an equipment control plan, including:

[0013] Using a convolutional neural network to extract features from the image data in the comprehensive data set to obtain image feature data;

[0014] Modeling the time series data in the comprehensive data set using a long short-term memory network to obtain time series feature data;

[0015] Combining the image feature data with the time series feature data to perform multi-source data fusion to obtain an enhanced comprehensive data set;

[0016] Using an integrated learning method to perform multi-model fusion on the enhanced comprehensive data set to generate a power demand forecast result;

[0017] Combining the power demand forecast results with real-time power market price information, a genetic algorithm is used to optimize the startup sequence, startup time and power output of different power-consuming devices to generate a device control plan including priority sorting.

[0018] Optionally, combining the power demand forecast results with real-time power market price information, using a genetic algorithm, the startup sequence, startup time and power output of different power-consuming devices are optimized to generate a device control plan with priority sorting, including:

[0019] Combining the power demand forecast result with the real-time power market electricity price information to perform economic evaluation and obtain an economic evaluation result;

[0020] Using a genetic algorithm to preliminarily optimize the startup sequence, startup time and power output of different electrical equipment in the economic evaluation results to obtain a preliminary optimization plan;

[0021] Introducing an interaction model between devices to simulate the device startup sequence, startup time and power output in the preliminary optimization scheme, and obtaining an optimization scheme for calculating the interaction between devices;

[0022] The optimization scheme is combined with the shelter safety operation standard to perform safety verification processing to obtain a safety verification result that meets safety specifications. Based on the safety verification result, an equipment control plan including priority sorting is generated.

[0023] Optionally, a device interaction model is introduced to simulate the device startup sequence, startup time and power output in the preliminary optimization scheme to obtain an optimization scheme for computing the interaction between devices, including:

[0024] Using the device interaction model to simulate the device startup sequence, startup time and power output in the optimization scheme, to obtain an initial device interaction result;

[0025] Applying a multi-objective optimization algorithm to adjust the initial device interaction result, calculate device energy consumption, startup time delay and system stability, adjust the startup interval and power allocation ratio between devices, and obtain an optimized device interaction result;

[0026] The optimized equipment interaction result is combined with the optimization scheme to conduct a comprehensive evaluation, and the environmental factor impact evaluation is introduced to obtain a comprehensive evaluation result;

[0027] An iterative optimization strategy is adopted in combination with a machine learning algorithm to automatically adjust the optimization parameters in the comprehensive evaluation results, and the simulation of the device interaction model and the adjustment of the multi-objective optimization algorithm are repeatedly performed until the preset optimization termination condition is reached, thereby obtaining an optimization solution for the interaction between computing devices.

[0028] Optionally, based on the equipment control plan, control instructions are output to the intelligent power distribution module in the cabin through the 5G network to achieve accurate power supply management of each electrical equipment, including:

[0029] Based on the device control plan, generate specific control instructions for each electrical device, including start-up time, power output level and operation mode;

[0030] The specific control instructions are sent to the intelligent power distribution module in the cabin through the 5G network to achieve precise control of the startup time, power output level and operation mode of each electrical device;

[0031] The intelligent power distribution module is used to monitor the actual operation of each electrical device in real time and collect operation status data, including current, voltage and power factor;

[0032] Compare and analyze the operating status data with the preset parameters in the equipment control plan to identify equipment with deviations;

[0033] For the device with deviation, dynamically adjust the control instruction of the online adaptive control algorithm to obtain a dynamic control instruction;

[0034] The dynamic control instructions are sent to the intelligent power distribution module through the 5G network to achieve precise power supply management for each electrical equipment.

[0035] Optionally, the operation data in the precise power supply management process is simulated using reinforcement learning technology to generate a new control strategy, including:

[0036] Reinforcement learning technology is used to simulate and process the operating data collected during the precise power supply management process, so as to construct a virtual environment for the operation of the power system in the shelter;

[0037] In the virtual environment, a reward function is defined to evaluate the system operation status under different control strategies, and evaluation results of multiple control strategies are obtained;

[0038] According to the evaluation results, exploring and selecting the best performing control strategy as a candidate strategy through a trial-and-error learning mechanism;

[0039] The candidate strategy is compared and verified with the existing control strategy to obtain a verification result. Based on the verification result, an optimized candidate strategy is generated through a preset number of iterative optimizations. According to the optimized candidate strategy, a new control strategy is generated.

[0040] Optionally, the operating status of the power system in the shelter is continuously monitored through the 5G network, actual operating data is collected, the actual operating data is combined with the new control strategy, the control strategy is adjusted, and an intelligent power supply control solution is formed, including:

[0041] Through the 5G network, the operating status of the power system in the shelter is continuously monitored and actual operating data is collected;

[0042] Combining the actual operation data with the new control strategy, performing real-time comparative analysis, identifying the difference between the operation status and the control strategy, and obtaining a difference analysis result;

[0043] Based on the difference analysis result, the control strategy is adjusted by using an adaptive learning algorithm to obtain an adjusted control strategy;

[0044] Applying the adjusted control strategy to a simulated environment of the electric power system in the shelter, performing simulation test verification, and obtaining verification results;

[0045] According to the verification result, the control strategy is optimized to obtain an optimized control strategy, and the optimized control strategy is deployed to the intelligent power control system in the cabin to form an intelligent power control solution.

[0046] In a second aspect, the embodiment of the present application provides a cabin intelligent power supply control system based on 5G communication, including:

[0047] An integration module, which uses multi-source data fusion technology to integrate the working status information, power demand signal and external environment data of the shelter to obtain a comprehensive data set;

[0048] An analysis module, which uses a deep learning algorithm to analyze the comprehensive data set to generate a power demand forecast result, combines the power demand forecast result with real-time power market electricity price information, and uses a genetic algorithm to optimize the start-up time and power output of different electrical equipment in the shelter to form an equipment control plan;

[0049] The output module outputs control instructions to the intelligent power distribution module in the cabin through the 5G network based on the equipment control plan, so as to realize accurate power supply management of each electrical equipment;

[0050] A simulation module, which uses reinforcement learning technology to simulate the operating data in the precise power supply management process to generate a new control strategy;

[0051] The adjustment module continuously monitors the operating status of the power system in the cabin through the 5G network, collects actual operating data, combines the actual operating data with the new control strategy, adjusts the control strategy, and forms an intelligent power supply control solution.

[0052] In a third aspect, an embodiment of the present invention provides a computing device, comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute a method for controlling an intelligent power supply of a cabin based on 5G communication as described in any one of the first aspects.

[0053] In a fourth aspect, an embodiment of the present invention provides a computer storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement a method for controlling an intelligent power supply of a cabin based on 5G communication as described in any one of the first aspects.

[0054] In the embodiment of the present invention, the working status information and power demand signal of each electrical device in the shelter and the external environment data of the shelter are received in real time through the 5G network;

[0055] The working status information, power demand signal and external environment data of the shelter are integrated by using multi-source data fusion technology to obtain a comprehensive data set;

[0056] A deep learning algorithm is used to analyze the comprehensive data set to generate a power demand forecast result, and the power demand forecast result is combined with real-time power market electricity price information, and a genetic algorithm is used to optimize the start-up time and power output of different electrical equipment in the cabin to form an equipment control plan;

[0057] Based on the equipment control plan, control instructions are output to the intelligent power distribution module in the cabin through the 5G network to achieve accurate power supply management for each electrical equipment;

[0058] Using reinforcement learning technology to simulate the operating data in the precise power supply management process to generate a new control strategy;

[0059] The operating status of the power system in the cabin is continuously monitored through the 5G network, actual operating data is collected, the actual operating data is combined with the new control strategy, the control strategy is adjusted, and an intelligent power supply control solution is formed.

[0060] The technical solution provided by the present invention improves the power management efficiency and intelligence level of the shelter facilities, and also significantly enhances the reliability and economy of the system, providing a more efficient and safe power guarantee solution for mobile medical care, emergency rescue and other fields. Among them, multi-objective optimization: the multi-objective optimization algorithm is applied to adjust the initial results of the interaction between devices, and multiple objective functions such as energy consumption, startup time delay and system stability are comprehensively considered. This method can find the best balance between multiple performance indicators, so as to achieve the optimal configuration of the overall system; comprehensive comprehensive evaluation: the optimized results of the interaction between devices are combined with the optimization scheme for comprehensive evaluation, and the influence of environmental factors is taken into account to ensure that the final solution meets the performance requirements while also meeting the specific needs of the actual application scenario.

[0061] These and other aspects of the present invention will become more apparent from the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0063] Figure 1 A flowchart of a method for controlling a modular cabin intelligent power supply based on 5G communication provided by an embodiment of the present invention;

[0064] Figure 2 A schematic diagram of the structure of a shelter intelligent power supply control system based on 5G communication provided in an embodiment of the present invention;

[0065] Figure 3 A schematic diagram of the structure of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0066] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0067] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0068] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0069] The present invention provides a method for controlling a shelter intelligent power supply based on 5G communication. Figure 1 ,include:

[0070] Step 101: Receive working status information and power demand signals of various electrical equipment in the shelter and external environment data of the shelter in real time through the 5G network;

[0071] In this step, the working status information refers to whether the equipment is running, the operating efficiency, etc., the power demand signal refers to the current or future power demand of the equipment, and the external environment data refers to environmental factors such as temperature, humidity, and light intensity that may affect the performance of the equipment.

[0072] For example, it receives the working status and power requirements of multiple life support systems inside the mobile medical cabin, such as ventilators, ECG monitors, etc., sent by built-in sensors, and at the same time receives real-time weather data from the meteorological station outside the cabin. All this information is quickly transmitted to the cloud processing center through the 5G network.

[0073] Step 102: Integrate the working status information, power demand signal and cabin external environment data using multi-source data fusion technology to obtain a comprehensive data set;

[0074] In this step, multi-source data fusion refers to the integration of data from different sources to obtain more comprehensive information. This process usually includes operations such as data cleaning, conversion, standardization and merging, aiming to eliminate redundancy, fill missing values, and improve data consistency and accuracy.

[0075] For example, integrating the mobile medical shelter mentioned above, all the received data is sent to a data fusion platform. The platform first cleans up invalid or erroneous data records, then unifies the different types of data into a unified format, and finally merges them into a complete data set. For example, the status information of a ventilator is associated with the temperature data at the same time point to form a new record.

[0076] Step 103: Analyze the comprehensive data set using a deep learning algorithm to generate a power demand forecast result, combine the power demand forecast result with real-time power market electricity price information, and use a genetic algorithm to optimize the start-up time and power output of different electrical equipment in the cabin to form an equipment control plan;

[0077] In this step, deep learning algorithm refers to a machine learning method based on neural networks, which is good at processing large amounts of complex data and can be used for various tasks such as pattern recognition, classification, and regression; genetic algorithm refers to a heuristic search algorithm that imitates natural selection and genetics mechanisms and is used to solve optimization problems.

[0078] For example, the comprehensive data set is analyzed and a deep learning model, such as a recurrent neural network (RNN), is used to predict the total power demand in the next few hours. At the same time, after obtaining real-time electricity price information, the genetic algorithm calculates the optimal equipment startup sequence and time, power setting and other parameters based on the predicted demand curve and current electricity price to ensure the lowest cost and meet demand.

[0079] Step 104: Based on the equipment control plan, control instructions are output to the intelligent power distribution module in the cabin through the 5G network to achieve accurate power supply management for each electrical device;

[0080] In this step, the intelligent power distribution module is responsible for adjusting the actual power supply of each device according to the received control instructions. This step ensures that the theoretical optimization solution can be actually applied on site.

[0081] For example, the best equipment control strategy is determined, and specific control instructions are quickly sent to the intelligent power management system in the cabin through the 5G network. For example, the power output of non-critical loads is increased during periods of low electricity prices, while the startup of certain equipment is reduced or delayed during peak hours.

[0082] Step 105: using reinforcement learning technology to simulate the operation data in the precise power supply management process to generate a new control strategy;

[0083] In this step, reinforcement learning is a method that allows machines to learn how to make decisions through trial and error, with the goal of maximizing some form of cumulative reward. In this case, it helps the system learn from actual operations and improve its control logic.

[0084] For example, after a period of operation, the collected operation data is used to train a reinforcement learning model. The model evaluates the effects of various control actions and adjusts future control strategies accordingly. For example, if it is found that power shortages occur frequently during a certain period of time, the model may recommend turning on the backup generator in advance.

[0085] Step 106: Continuously monitor the operating status of the power system in the cabin through the 5G network, collect actual operating data, combine the actual operating data with the new control strategy, adjust the control strategy, and form an intelligent power supply control solution;

[0086] In this step, this step emphasizes the system's continuous learning and self-optimization capabilities. Through the continuous collection and analysis of actual operating data, the system can dynamically adjust its control strategy to adapt to changing conditions.

[0087] For example, as the seasons change or other external factors change, the power demand pattern in the cabin will also change. The data continuously collected through the 5G network is used to update the existing reinforcement learning model to generate a new control strategy that is more suitable for the current situation. This iterative optimization process ensures that the entire system is always in the best operating state.

[0088] The present invention provides a specific embodiment, wherein the step 103 uses a deep learning algorithm to analyze the comprehensive data set to generate a power demand forecast result, combines the power demand forecast result with real-time power market electricity price information, and uses a genetic algorithm to optimize the start-up time and power output of different electrical equipment in the shelter to form an equipment control plan, which specifically includes the following steps:

[0089] Step 301: extracting features of image data in the comprehensive data set using a convolutional neural network to obtain image feature data;

[0090] In this step, convolutional neural network refers to a deep learning model that is particularly suitable for processing data with spatial structures, such as images. The convolutional neural network automatically extracts key features in the image through operations such as convolution layers and pooling layers. Image data refers to visual information inside and outside the cabin, such as equipment status indicator lights and images captured by environmental monitoring cameras.

[0091] For example, to extract image data from a medical cabin, a camera can be used to monitor the status indicator lights and display screens of the equipment. These image data are processed through a convolutional neural network to extract key features that represent the operating status of the equipment. For example, parameter changes on the ventilator display can be identified to determine whether it is working properly.

[0092] Step 302: Use the long short-term memory network to model the time series data in the comprehensive data set to obtain time series feature data

[0093] In this step, the long short-term memory network refers to a variant of a recursive neural network, which is particularly suitable for processing long sequence data and can capture long-term dependencies in time series; time series data refers to changes in the working status of equipment over time, historical records of power demand, time changes in external environmental parameters, etc.

[0094] For example, collect the working status data of each device in the cabin over a period of time, such as hourly power consumption, temperature changes, etc.; use the long short-term memory network to model these time series data and extract features that can reflect the equipment's operating mode and trend, such as predicting power demand in the next few hours.

[0095] Step 303: Combine the image feature data with the time series feature data to perform multi-source data fusion to obtain an enhanced comprehensive data set

[0096] In this step, multi-source data fusion refers to integrating data from different sources to obtain more comprehensive information. This step usually involves operations such as data alignment, feature splicing, or feature-level fusion.

[0097] For example, combining equipment status features extracted from image data and power demand features extracted from time series data, such as combining parameter changes on a ventilator display with power consumption data from the past few hours, forms a dataset with more contextual information for subsequent analysis.

[0098] Step 304: Use an ensemble learning method to perform multi-model fusion on the enhanced comprehensive data set to generate a power demand forecast result

[0099] In this step, ensemble learning refers to improving prediction performance by combining multiple machine learning models. Common methods include bagging, boosting, and stacking. Multi-model fusion refers to using a variety of different models, such as decision trees, support vector machines, neural networks, etc., for training and integrating their prediction results.

[0100] For example, ensemble learning methods, such as random forests and gradient boosting trees, are used to train enhanced comprehensive data sets. Each model extracts different features from the data and makes predictions based on its characteristics, and ultimately generates more accurate electricity demand forecast results through voting or weighted averaging.

[0101] Step 305: Combine the power demand forecast results with the real-time power market price information, use genetic algorithms to optimize the startup sequence, startup time and power output of different power-consuming devices, and generate a device control plan with priority sorting.

[0102] In this step, genetic algorithm refers to a heuristic search algorithm that imitates natural selection and genetics mechanisms to solve complex optimization problems. Priority sorting refers to assigning startup sequence, startup time and power output to each device based on the optimization results to ensure that the system meets the needs while minimizing the cost.

[0103] For example, by combining power demand forecast results and real-time electricity price information, genetic algorithms are used to optimize the startup sequence and power settings of each device in the cabin. For example, high-energy-consuming equipment is started first during periods when electricity prices are lower, while the use of non-critical equipment is reduced during peak hours. Ultimately, an equipment control plan with detailed priority sorting is generated to ensure the efficient operation of the entire system.

[0104] The present invention provides a specific embodiment, wherein the step 305 combines the power demand forecast result with the real-time power market price information, uses a genetic algorithm to optimize the startup sequence, startup time and power output of different power-consuming devices, and generates a device control plan including priority sorting, specifically comprising the following steps:

[0105] Step 311: Combine the power demand forecast result with the real-time power market electricity price information to perform economic evaluation and obtain an economic evaluation result.

[0106] In this step, economic evaluation refers to calculating the cost-effectiveness of different electricity usage strategies by analyzing electricity demand forecasts and real-time electricity price information, which includes considering the impact of equipment startup sequence, startup time and power output on the total electricity bill; real-time electricity market electricity price information refers to real-time price fluctuation data in the electricity market, which is used to determine the electricity cost in different time periods.

[0107] For example, in a mobile medical cabin, assuming that the power demand for the next 24 hours has been predicted by a deep learning model, and the hourly electricity price information has been obtained from the power market, this information can be combined to calculate the cost of starting specific equipment, such as CT scanners, ventilators, etc., during different time periods. For example, if it is predicted that the power demand is lower and the electricity price is cheaper at night, high-energy-consuming equipment can be prioritized to operate at night to reduce overall electricity bill expenditure.

[0108] Step 312: Using a genetic algorithm to preliminarily optimize the startup sequence, startup time and power output of different electrical equipment in the economic evaluation results to obtain a preliminary optimization solution

[0109] In this step, genetic algorithm refers to a heuristic search algorithm that imitates natural selection and genetics mechanisms to solve complex optimization problems. It generates new solutions through selection, crossover and mutation operations and gradually improves them. The preliminary optimization plan refers to finding a preliminary equipment scheduling plan based on the economic evaluation results through genetic algorithm, which minimizes costs while meeting demand.

[0110] For example, the above economic evaluation results are processed using a genetic algorithm to generate a preliminary equipment scheduling plan. For example, the algorithm may recommend starting a high-energy consumption CT scanner during the period when electricity prices are the lowest, and reducing the use of non-critical equipment during periods when electricity prices are higher. Through multiple iterations, the algorithm will gradually optimize the equipment's startup sequence, startup time, and power output to achieve the best economic benefits.

[0111] Step 313: Introduce the device interaction model to simulate the device startup sequence, startup time and power output in the preliminary optimization scheme to obtain an optimization scheme for computing the interaction between devices.

[0112] In this step, the device interaction model refers to a mathematical or simulation model used to describe and predict the mutual influence between different devices, including the physical connection, energy flow and possible interference between devices; the optimization scheme for calculating the interaction between devices refers to further adjusting the preliminary optimization scheme by simulating the interaction between devices to ensure that the devices can work together to avoid conflicts and inefficiency.

[0113] For example, based on the preliminary optimization plan, a model of interaction between devices is introduced to simulate the situation when different devices are started at the same time. For example, if multiple large devices are started at the same time, which may cause power grid overload, the model will readjust the startup sequence and time interval of the devices to ensure power grid stability. In addition, the model will also consider the dependencies between devices, such as some devices need to be started before other devices can run, thereby generating a more reasonable optimization plan.

[0114] Step 314: Combine the optimization scheme with the shelter safety operation standard to perform safety verification processing to obtain safety verification results that meet safety standards. Based on the safety verification results, generate an equipment control plan with priority sorting.

[0115] In this step, safety verification refers to checking whether the optimization plan meets the preset safety standards and specifications to ensure that the equipment operation will not cause safety hazards; safety verification results refer to the results after safety verification, indicating whether the optimization plan meets all safety requirements; the prioritized equipment control plan refers to the final generated equipment scheduling plan, which includes the equipment startup sequence, startup time and power output, and is sorted by priority to ensure that the system runs efficiently under the premise of safety.

[0116] For example, the above optimization plan is compared with the safe operation standards of the cabin to ensure that the operation of all equipment is within the safe range. For example, check whether the maximum power output of the equipment exceeds the carrying capacity of the power grid, and confirm that the startup sequence of the equipment will not cause power grid instability. After passing the safety verification, generate the final equipment control plan, which includes the specific startup time, power setting and priority ranking of each device. For example, life support systems, such as ventilators, will be given the highest priority to ensure that they are always in the best operating state.

[0117] Based on this, the present invention provides a specific embodiment, wherein the step 313 introduces an interaction model between devices to simulate the device startup sequence, startup time and power output in the preliminary optimization scheme to obtain an optimization scheme for computing the interaction between devices, specifically comprising the following steps:

[0118] Step 321: Use the device interaction model to simulate the device startup sequence, startup time and power output in the optimization scheme to obtain an initial device interaction result.

[0119] In this step, the device interaction model refers to a mathematical or simulation model used to describe and predict the mutual influence between different devices, including the physical connection, energy flow, electromagnetic interference and possible dependencies between devices; the initial device interaction results refer to the preliminary results obtained through model simulation, which show the interaction between devices under a specific startup sequence, startup time and power output.

[0120] For example, in a mobile medical cabin, assuming that a preliminary equipment control plan has been obtained, the equipment interaction model is used to simulate the situation when these devices are started according to the plan. For example, the model can predict whether the power grid will be overloaded or the voltage fluctuates when the CT scanner and X-ray machine are started at the same time. The model can also consider the dependencies between devices, such as some diagnostic equipment needs to be started after the life support system is running stably.

[0121] Step 322: Apply a multi-objective optimization algorithm to adjust the initial device interaction result, calculate device energy consumption, startup time delay, and system stability, adjust the startup interval and power allocation ratio between devices, and obtain an optimized device interaction result.

[0122] In this step, the multi-objective optimization algorithm refers to an optimization method that solves multiple objective functions, such as minimizing cost, maximizing efficiency, etc. Common multi-objective optimization algorithms include NSGA-II, MOEA / D, etc.; the optimized device interaction result refers to the better device startup sequence, startup time and power output solution obtained after adjustment by the multi-objective optimization algorithm.

[0123] For example, based on the initial device interaction results, a multi-objective optimization algorithm, such as NSGA-II, is applied to adjust the startup sequence, startup time, and power output of the devices. The algorithm will comprehensively consider multiple objectives such as device energy consumption, startup time delay, and system stability. For example, by adjusting the startup interval and power allocation ratio of the devices, it can ensure that the power grid will not be overloaded, while minimizing the startup time delay of the devices and maintaining the stable operation of the system.

[0124] More specifically, the present invention also provides a formula to calculate the optimized device interaction result, and the specific calculation formula is as follows:

[0125] Objective Function = w 1 ·E+w 2 ·T+w 3 ·S+w4 ·C

[0126] Among them, Objective Function is a core concept in mathematical optimization problems, which is used to quantify and measure the objectives or target values ​​in optimization problems;

[0127] The calculation formula of equipment energy consumption is:

[0128]

[0129] Where E represents the energy consumption of the equipment; P i is the power output of the ith device; t i is the running time of the i-th device; N is the total number of devices;

[0130] The calculation formula for the startup time delay is:

[0131]

[0132] Where T represents the start-up time delay; t start,i is the actual startup time of the ith device; t desired,i is the expected startup time of the i-th device; S represents the system stability, which is calculated as:

[0133]

[0134] ΔV i is the voltage fluctuation of the ith device; V i is the rated voltage of the i-th device;

[0135] The calculation formula for the coordination between devices is:

[0136]

[0137] Among them, C represents the coordination between devices; P i and t i are the power output and operating time of the i-th device respectively;

[0138] P j and t j are the power output and operating time of the jth device respectively;

[0139] w 1 ,w 2 ,w 3 ,w 4 They are the weight coefficients of equipment energy consumption, startup time delay, system stability and coordination between equipment, satisfying w 1 +w 2 +w 3 +w 4=1.

[0140] Step 323: Combine the optimized device interaction results with the optimization plan to conduct a comprehensive evaluation, and introduce the environmental factors to evaluate the impact, to obtain a comprehensive evaluation result.

[0141] In this step, comprehensive evaluation refers to a comprehensive performance evaluation based on the optimized equipment interaction results and the overall optimization plan, which includes considerations of economy, efficiency, safety, and other aspects. Environmental factor impact assessment refers to considering the impact of external environmental factors such as temperature, humidity, and light on equipment operation to ensure that the optimization plan is still effective in the actual environment.

[0142] For example, evaluate the operating efficiency and stability of equipment under different temperature and humidity conditions. If it is found that the performance of some equipment decreases under high temperature environment, additional cooling measures can be added to the optimization plan or the operating parameters of the equipment can be adjusted. The comprehensive evaluation results will provide a basis for the final equipment control plan.

[0143] Step 324: Adopt an iterative optimization strategy and combine it with a machine learning algorithm to automatically adjust the optimization parameters in the comprehensive evaluation result, repeatedly perform the simulation of the device interaction model and the adjustment of the multi-objective optimization algorithm until the preset optimization termination condition is reached, and obtain an optimization solution for computing the interaction between devices

[0144] In this step, the iterative optimization strategy refers to continuously improving the optimization plan through multiple iterations until the preset optimization termination conditions are met; the machine learning algorithm refers to using historical data and current evaluation results to automatically adjust the optimization parameters to improve the optimization efficiency and effect.

[0145] For example, through reinforcement learning algorithms, based on the feedback of the results of each iteration, the startup sequence, startup time and power output of the equipment are gradually adjusted to achieve the best economic benefits and system stability. The iterative process continues until the preset optimization termination conditions are reached, such as the total energy consumption is reduced below a certain threshold, or the system stability reaches the predetermined standard. The final optimization plan will be a highly adaptive and efficient equipment control plan.

[0146] Based on this, the present invention provides a specific embodiment, wherein step 104 outputs control instructions to the intelligent power distribution module in the cabin through the 5G network based on the equipment control plan to achieve accurate power supply management of each electrical device, specifically including the following steps:

[0147] Step 401: Based on the device control plan, generate specific control instructions for each electrical device, including start-up time, power output level and operation mode;

[0148] In this step, the device control plan refers to a detailed scheduling plan, which includes the startup sequence, startup time, power output and operation mode of each device; the specific control instructions refer to the actual operation instructions generated according to the device control plan, which are used to instruct the intelligent power distribution module how to control each device.

[0149] For example, in a mobile medical cabin, assuming that an equipment control plan has been developed, the plan may stipulate that the CT scanner starts at 8:00 a.m., runs at 70% power, and is set to diagnostic mode; at the same time, the ventilator starts at 7:30 a.m., runs at 50% power, and is set to standard mode. Based on this plan, the system will generate specific control instructions such as "CT scanner: start at 8:00, 70% power, diagnostic mode" and "ventilator: start at 7:30, 50% power, standard mode".

[0150] Step 402: Send the specific control instruction to the intelligent power distribution module in the cabin through the 5G network to achieve accurate control of the startup time, power output level and operation mode of each electrical device;

[0151] In this step, 5G network refers to the fifth generation of mobile communication technology, which provides high-speed, low-latency and high-reliability data transmission; intelligent power distribution module refers to the hardware or software system responsible for receiving control instructions and performing corresponding operations to ensure that the equipment operates in the intended manner.

[0152] For example, by utilizing the high speed and low latency characteristics of the 5G network, the above-mentioned specific control instructions are quickly sent to the intelligent power distribution module in the cabin. For example, when the control instruction of the CT scanner arrives, the intelligent power distribution module will start the CT scanner on time at 8:00 and set its power to 70%, and switch to diagnostic mode at the same time. Similarly, the ventilator will also start on time at 7:30 and be set to 50% power and standard mode.

[0153] Step 403: using the intelligent power distribution module to monitor the actual operation of each electrical device in real time and collect operation status data, wherein the operation status data includes current, voltage and power factor;

[0154] In this step, real-time monitoring refers to continuously monitoring the operating status of the equipment in order to detect abnormal conditions in a timely manner; operating status data refers to data reflecting the actual working conditions of the equipment, such as current, voltage, and power factor.

[0155] For example, the intelligent power distribution module has built-in sensors that can monitor the current, voltage and power factor of the CT scanner and ventilator in real time. For example, the current of the CT scanner is 10A, the voltage is 220V, and the power factor is 0.95; the current of the ventilator is 5A, the voltage is 220V, and the power factor is 0.90. These data will be continuously recorded and stored for subsequent analysis.

[0156] Step 404: comparing and analyzing the operating status data with the preset parameters in the equipment control plan to identify equipment with deviations;

[0157] In this step, comparative analysis refers to comparing the actual operating status data with the preset parameters to find out the inconsistencies; deviation refers to the difference between the actual operating status and the preset parameters, which may be caused by equipment failure, changes in the external environment or other factors.

[0158] For example, the system compares the real-time collected operating status data with the preset parameters in the equipment control plan. For example, if the preset current of the CT scanner is 10A, but the actual monitored current is 12A, this indicates a deviation. Similarly, if the preset power factor of the ventilator is 0.90, but the actual value is 0.85, this is also a deviation.

[0159] Step 405: For the device with deviation, dynamically adjust the control instruction of the online adaptive control algorithm to obtain a dynamic control instruction;

[0160] In this step, the online adaptive control algorithm refers to an algorithm that can automatically adjust the control strategy according to real-time data to cope with changes in the system, and the dynamic control instructions refer to new control instructions generated according to the online adaptive control algorithm to correct deviations and optimize equipment operation.

[0161] For example, for detected deviations, the system uses an online adaptive control algorithm to generate new control instructions. For example, if the current of a CT scanner is too high, the algorithm may recommend reducing its power output to 60%. For a ventilator, if the power factor is too low, the algorithm may recommend adjusting its operating mode or increasing compensation capacitance to improve the power factor. These newly generated control instructions are dynamic control instructions.

[0162] Step 406: Send the dynamic control instruction to the intelligent power distribution module through the 5G network to achieve accurate power supply management for each power-consuming device;

[0163] In this step, dynamic control instructions refer to new control instructions generated based on real-time data analysis, which are used to correct deviations and optimize equipment operation; precise power supply management refers to dynamically adjusting control instructions to ensure that the equipment is always in the best operating state and improve the overall efficiency and stability of the system.

[0164] For example, the generated dynamic control instructions are quickly sent to the intelligent power distribution module through the 5G network. For example, after the intelligent power distribution module receives the instruction to reduce the power output of the CT scanner to 60%, it will immediately execute the operation. For the ventilator, if it receives the instruction to increase the compensation capacitor, the intelligent power distribution module will automatically adjust the relevant settings to improve its power factor. The entire system can respond to changes in real time to ensure that all equipment operates efficiently and stably.

[0165] Based on this, the present invention provides a specific embodiment, in which step 105 uses reinforcement learning technology to simulate the operating data in the precise power supply management process to generate a new control strategy, which specifically includes the following steps:

[0166] Step 501: using reinforcement learning technology to simulate and process the operation data collected during the precise power supply management process, and constructing a virtual environment for the operation of the power system in the shelter;

[0167] In this step, reinforcement learning technology refers to a machine learning method that learns how to make decisions through the interaction between the agent and the environment. The agent obtains rewards or penalties through a trial and error process and gradually optimizes its behavior strategy; the virtual environment refers to a simulation model built based on actual operating data to simulate the operation of the real system. The virtual environment can be used to test different control strategies without implementing them in the actual system.

[0168] For example, in a mobile medical cabin, a virtual environment is constructed using the collected operating data, such as current, voltage, power factor, etc. This virtual environment can simulate the operating status of the power system in the cabin, including the start-up time, power output and operating mode of each device. For example, a reinforcement learning model can be trained using historical data, which can simulate the system response under different control strategies.

[0169] Step 502: In the virtual environment, a reward function is defined to evaluate the system operating status under different control strategies to obtain evaluation results of multiple control strategies;

[0170] In this step, the reward function refers to the function used to measure the behavior of the intelligent agent in reinforcement learning. The design of the reward function directly affects the learning direction of the intelligent agent and the quality of the final strategy; the evaluation result refers to the performance indicators of different control strategies in the virtual environment calculated by the reward function, such as energy consumption, system stability, cost-effectiveness, etc.

[0171] For example, in a virtual environment, a reward function is defined that takes into account multiple objectives, such as minimizing total energy consumption, maximizing system stability, minimizing operating costs, etc. For example, a control strategy may perform well in terms of energy consumption but poorly in terms of system stability, and ultimately its overall performance is comprehensively evaluated through the reward function.

[0172] Step 503: Based on the evaluation results, explore and select the best performing control strategy as a candidate strategy through a trial-and-error learning mechanism;

[0173] In this step, the trial-and-error learning mechanism refers to the core mechanism of reinforcement learning, which gradually finds the optimal solution by constantly trying different actions and adjusting the strategy based on reward feedback; the candidate strategy refers to the control strategy that performs best after multiple trial-and-error learning.

[0174] For example, in a virtual environment, multiple trials are conducted through reinforcement learning algorithms such as Q-learning, Deep Q-Network (DQN), etc., and different control strategies are adopted in each trial, and the corresponding reward values ​​are recorded. By comparing these reward values, the best performing control strategy is selected as the candidate strategy. For example, if a certain strategy makes the system have the lowest energy consumption and the highest stability in multiple trials, then this strategy will be selected as the candidate strategy.

[0175] Step 504: Compare and verify the candidate strategy with the existing control strategy to obtain a verification result, and based on the verification result, generate an optimized candidate strategy through a preset number of iterative optimizations, and generate a new control strategy based on the optimized candidate strategy;

[0176] In this step, comparative verification processing refers to comparing the candidate strategy with the existing control strategy under the same conditions to verify the superiority of the candidate strategy; iterative optimization refers to continuously improving the candidate strategy through multiple iterations until the preset optimization termination conditions are reached; the new control strategy refers to the final control strategy after optimization, which is used in the actual system.

[0177] For example, the candidate strategy is compared with the existing control strategy in a virtual environment for verification. For example, assuming that the average energy consumption of the existing control strategy is 100kWh / day and the system stability is 95%, while the average energy consumption of the candidate strategy is 85kWh / day and the system stability is 98%. Through comparative verification, it is confirmed that the candidate strategy is superior to the existing strategy in all aspects.

[0178] Based on this, the present invention provides a specific embodiment, wherein step 106 continuously monitors the operating status of the power system in the cabin through the 5G network, collects actual operating data, combines the actual operating data with the new control strategy, adjusts the control strategy, and forms an intelligent power supply control solution, which specifically includes the following steps:

[0179] Step 601: Continuously monitor the operating status of the power system in the shelter through the 5G network and collect actual operating data;

[0180] In this step, 5G network refers to the fifth generation of mobile communication technology, which provides high-speed, low-latency and high-reliability data transmission; actual operation data refers to real-time data including current, voltage, power factor, equipment status, etc., which are used to reflect the current operating status of the power system.

[0181] For example, in a mobile medical cabin, various sensors and monitoring devices connected by 5G networks, such as current sensors, voltage sensors, power factor meters and other tools, continuously collect operating data of the power system. These data include the real-time current, voltage, power factor and working status of each electrical device. For example, the current of the CT scanner is 10A, the voltage is 220V, and the power factor is 0.95; the current of the ventilator is 5A, the voltage is 220V, and the power factor is 0.90. These data are transmitted to the central control system in real time through the 5G network.

[0182] Step 602: combining the actual operation data with the new control strategy, performing real-time comparative analysis, identifying the difference between the operation state and the control strategy, and obtaining a difference analysis result;

[0183] In this step, real-time comparative analysis refers to comparing actual operating data with preset control strategies to identify the differences between the two; difference analysis results refer to the results obtained through comparative analysis, showing the deviation between the actual operating status and the control strategy.

[0184] For example, in the central control system, the actual operating data collected is compared with the newly formulated control strategy in real time. For example, if the control strategy requires the CT scanner to start at 8:00 and run at 70% power, but the actual monitoring data shows that the CT scanner is not started until 8:05 and the power output is 65%, there is a time delay and power output deviation. The system will record these differences and generate a difference analysis report.

[0185] Step 603: Based on the difference analysis result, the control strategy is adjusted using an adaptive learning algorithm to obtain an adjusted control strategy;

[0186] In this step, the adaptive learning algorithm refers to an algorithm that can automatically adjust the control parameters based on real-time feedback to adapt to system changes; the adjusted control strategy refers to the new control strategy optimized by the adaptive learning algorithm, which aims to reduce differences and improve system performance.

[0187] For example, based on the results of the difference analysis, adaptive learning algorithms such as recursive least squares, Kalman filter, etc. are used to adjust the control strategy. For example, if it is found that the startup time of the CT scanner is always later than expected, the algorithm may advance its startup time; if the power output is inconsistent, the algorithm may adjust the parameters of the power regulator. Through these adjustments, a new control strategy is generated to reduce the difference between the actual operating status and the control strategy.

[0188] Step 604: applying the adjusted control strategy to a simulated environment of the power system in the shelter, performing a simulation test verification, and obtaining a verification result;

[0189] In this step, the simulation environment refers to a simulation model built based on actual operating data, which is used to test and verify the effectiveness of the control strategy; the verification result refers to the result obtained through simulation testing, which evaluates the performance of the adjusted control strategy in the virtual environment.

[0190] For example, the adjusted control strategy is applied to the previously constructed virtual environment for simulation testing. For example, in the simulation environment, the startup time and power output of the CT scanner are set according to the new control strategy, and its operation in the virtual environment is observed. Through simulation testing, it can be verified whether the adjusted control strategy effectively reduces the startup time delay and power output deviation, and ensures system stability and efficiency.

[0191] Step 605: Optimize the control strategy according to the verification result to obtain the optimized control strategy, and deploy the optimized control strategy to the intelligent power control system in the cabin to form an intelligent power control solution.

[0192] In this step, the optimized control strategy refers to the control strategy that is further improved based on the simulation test verification results to achieve the best effect; the intelligent power supply control scheme refers to the finalized control strategy used to guide the operation of the actual system.

[0193] For example, based on the verification results of the simulation test, the control strategy is further optimized. For example, if the simulation test shows that the adjusted control strategy still has a small time delay, it can be further optimized by fine-tuning the algorithm parameters or adding redundant control logic. The optimized control strategy is deployed to the intelligent power control system in the cabin. For example, the startup time setting and power output control parameters of the CT scanner are updated to ensure that it is more accurate and efficient in actual operation, forming a complete intelligent power control solution to ensure the stable operation and efficient management of the power system in the cabin.

[0194] Figure 2 A structural diagram of a cabin intelligent power supply control system based on 5G communication is provided for the embodiment of the present application, such as Figure 2As shown, the system includes:

[0195] The receiving module 21 receives the working status information and power demand signal of each electrical device in the cabin and the external environment data of the cabin in real time through the 5G network;

[0196] An integration module 22 integrates the working status information, power demand signal and cabin external environment data using multi-source data fusion technology to obtain a comprehensive data set;

[0197] The analysis module 23 uses a deep learning algorithm to analyze the comprehensive data set to generate a power demand forecast result, combines the power demand forecast result with real-time power market electricity price information, and uses a genetic algorithm to optimize the start-up time and power output of different electrical equipment in the cabin to form an equipment control plan;

[0198] The output module 24 outputs control instructions to the intelligent power distribution module in the cabin through the 5G network based on the equipment control plan, so as to realize accurate power supply management of each electrical equipment;

[0199] A simulation module 25, using reinforcement learning technology to simulate the operation data in the precise power supply management process to generate a new control strategy;

[0200] The adjustment module 26 continuously monitors the operating status of the power system in the cabin through the 5G network, collects actual operating data, combines the actual operating data with the new control strategy, adjusts the control strategy, and forms an intelligent power supply control solution.

[0201] Figure 2 The 5G communication-based intelligent power supply control system for shelters can be implemented Figure 1 The implementation principle and technical effect of the xx method described in the embodiment shown will not be repeated. For the above-mentioned embodiment of a shelter intelligent power supply control system based on 5G communication, the specific way in which each module and unit performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0202] Figure 2 A 5G communication-based intelligent power supply control system for a shelter in the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0203] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0204] The processing component 32 is used to: receive the working status information and power demand signal of each electrical device in the cabin and the external environment data of the cabin in real time through the 5G network;

[0205] The working status information, power demand signal and external environment data of the shelter are integrated by using multi-source data fusion technology to obtain a comprehensive data set;

[0206] A deep learning algorithm is used to analyze the comprehensive data set to generate a power demand forecast result, and the power demand forecast result is combined with real-time power market electricity price information, and a genetic algorithm is used to optimize the start-up time and power output of different electrical equipment in the cabin to form an equipment control plan;

[0207] Based on the equipment control plan, control instructions are output to the intelligent power distribution module in the cabin through the 5G network to achieve accurate power supply management for each electrical equipment;

[0208] Using reinforcement learning technology to simulate the operating data in the precise power supply management process to generate a new control strategy;

[0209] The operating status of the power system in the cabin is continuously monitored through the 5G network, actual operating data is collected, the actual operating data is combined with the new control strategy, the control strategy is adjusted, and an intelligent power supply control solution is formed.

[0210] The processing component 32 includes one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component can also be implemented by one or more application-specific integrated circuits (AICs), digital signal processors (DPs), digital signal processing devices (DPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0211] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (RAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0212] Computing devices also include other components, such as input / output interfaces, display components, and communication components.

[0213] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above-mentioned peripheral interface module can be an output device or an input device.

[0214] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0215] Wherein, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device can refer to a cloud server, and the above-mentioned processing component, storage component, etc. can be basic server resources rented or purchased from a cloud computing platform.

[0216] An embodiment of the present invention also provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the above Figure 1 A 5G communication-based intelligent power control method and system for a mobile cabin shown in the embodiment.

[0217] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, device, and unit can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0218] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0219] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0220] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling an intelligent power supply of a shelter based on 5G communication, characterized in that: include: Receive the working status information and power demand signals of each electrical device in the shelter and the external environment data of the shelter in real time through the 5G network; The working status information, power demand signal and external environment data of the shelter are integrated by using multi-source data fusion technology to obtain a comprehensive data set; A deep learning algorithm is used to analyze the comprehensive data set to generate a power demand forecast result, and the power demand forecast result is combined with real-time power market electricity price information, and a genetic algorithm is used to optimize the start-up time and power output of different electrical equipment in the cabin to form an equipment control plan; Based on the equipment control plan, control instructions are output to the intelligent power distribution module in the cabin through the 5G network to achieve accurate power supply management for each electrical equipment; Using reinforcement learning technology to simulate the operating data in the precise power supply management process to generate a new control strategy; The operating status of the power system in the cabin is continuously monitored through the 5G network, actual operating data is collected, the actual operating data is combined with the new control strategy, the control strategy is adjusted, and an intelligent power supply control solution is formed.

2. The method according to claim 1, characterized in that The comprehensive data set is analyzed using a deep learning algorithm to generate a power demand forecast result. The power demand forecast result is combined with real-time power market price information, and a genetic algorithm is used to optimize the start-up time and power output of different electrical equipment in the shelter to form an equipment control plan, including: Using a convolutional neural network to extract features from the image data in the comprehensive data set to obtain image feature data; Modeling the time series data in the comprehensive data set using a long short-term memory network to obtain time series feature data; Combining the image feature data with the time series feature data to perform multi-source data fusion to obtain an enhanced comprehensive data set; Using an integrated learning method to perform multi-model fusion on the enhanced comprehensive data set to generate a power demand forecast result; Combining the power demand forecast results with real-time power market price information, a genetic algorithm is used to optimize the startup sequence, startup time and power output of different power-consuming devices to generate a device control plan including priority sorting.

3. The method according to claim 2, characterized in that Combining the power demand forecast results with the real-time power market price information, using genetic algorithms, the startup sequence, startup time and power output of different power-consuming devices are optimized to generate a device control plan with priority sorting, including: Combining the power demand forecast result with the real-time power market electricity price information to perform economic evaluation and obtain an economic evaluation result; Using a genetic algorithm to preliminarily optimize the startup sequence, startup time and power output of different electrical equipment in the economic evaluation results to obtain a preliminary optimization plan; Introducing an interaction model between devices to simulate the device startup sequence, startup time and power output in the preliminary optimization scheme, and obtaining an optimization scheme for calculating the interaction between devices; The optimization scheme is combined with the shelter safety operation standard to perform safety verification processing to obtain a safety verification result that meets safety specifications. Based on the safety verification result, an equipment control plan including priority sorting is generated.

4. The method according to claim 3, characterized in that The device interaction model is introduced to simulate the device startup sequence, startup time and power output in the preliminary optimization scheme to obtain an optimization scheme for computing the interaction between devices, including: Using the device interaction model to simulate the device startup sequence, startup time and power output in the optimization scheme, to obtain an initial device interaction result; Applying a multi-objective optimization algorithm to adjust the initial device interaction result, calculate device energy consumption, startup time delay and system stability, adjust the startup interval and power allocation ratio between devices, and obtain an optimized device interaction result; The optimized equipment interaction result is combined with the optimization scheme to conduct a comprehensive evaluation, and the environmental factor impact evaluation is introduced to obtain a comprehensive evaluation result; An iterative optimization strategy is adopted in combination with a machine learning algorithm to automatically adjust the optimization parameters in the comprehensive evaluation results, and the simulation of the device interaction model and the adjustment of the multi-objective optimization algorithm are repeatedly performed until the preset optimization termination condition is reached, thereby obtaining an optimization solution for the interaction between computing devices.

5. The method according to claim 1, characterized in that Based on the equipment control plan, control instructions are output to the intelligent power distribution module in the cabin through the 5G network to achieve precise power supply management for each electrical equipment, including: Based on the device control plan, generate specific control instructions for each electrical device, including start-up time, power output level and operation mode; The specific control instructions are sent to the intelligent power distribution module in the cabin through the 5G network to achieve precise control of the startup time, power output level and operation mode of each electrical device; The intelligent power distribution module is used to monitor the actual operation of each electrical device in real time and collect operation status data, including current, voltage and power factor; Compare and analyze the operating status data with the preset parameters in the equipment control plan to identify equipment with deviations; For the device with deviation, dynamically adjust the control instruction of the online adaptive control algorithm to obtain a dynamic control instruction; The dynamic control instructions are sent to the intelligent power distribution module through the 5G network to achieve precise power supply management for each electrical equipment.

6. The method according to claim 1, characterized in that The operation data in the precise power supply management process is simulated using reinforcement learning technology to generate a new control strategy, including: Reinforcement learning technology is used to simulate and process the operating data collected during the precise power supply management process, so as to construct a virtual environment for the operation of the power system in the shelter; In the virtual environment, a reward function is defined to evaluate the system operation status under different control strategies, and evaluation results of multiple control strategies are obtained; According to the evaluation results, exploring and selecting the best performing control strategy as a candidate strategy through a trial-and-error learning mechanism; The candidate strategy is compared and verified with the existing control strategy to obtain a verification result. Based on the verification result, an optimized candidate strategy is generated through a preset number of iterative optimizations. According to the optimized candidate strategy, a new control strategy is generated.

7. The method according to claim 1, characterized in that The operating status of the power system in the cabin is continuously monitored through the 5G network, actual operating data is collected, the actual operating data is combined with the new control strategy, the control strategy is adjusted, and an intelligent power supply control solution is formed, including: Through the 5G network, the operating status of the power system in the shelter is continuously monitored and actual operating data is collected; Combining the actual operation data with the new control strategy, performing real-time comparative analysis, identifying the difference between the operation status and the control strategy, and obtaining a difference analysis result; Based on the difference analysis result, the control strategy is adjusted by using an adaptive learning algorithm to obtain an adjusted control strategy; Applying the adjusted control strategy to a simulated environment of the electric power system in the shelter, performing simulation test verification, and obtaining verification results; According to the verification result, the control strategy is optimized to obtain an optimized control strategy, and the optimized control strategy is deployed to the intelligent power control system in the cabin to form an intelligent power control solution.

8. A shelter intelligent power supply control system based on 5G communication, characterized in that: include: The receiving module receives the working status information and power demand signals of each electrical device in the shelter and the external environment data of the shelter in real time through the 5G network; An integration module, which uses multi-source data fusion technology to integrate the working status information, power demand signal and external environment data of the shelter to obtain a comprehensive data set; An analysis module, which uses a deep learning algorithm to analyze the comprehensive data set to generate a power demand forecast result, combines the power demand forecast result with real-time power market electricity price information, and uses a genetic algorithm to optimize the start-up time and power output of different electrical equipment in the cabin to form an equipment control plan; The output module outputs control instructions to the intelligent power distribution module in the cabin through the 5G network based on the equipment control plan, so as to realize accurate power supply management of each electrical equipment; A simulation module, which uses reinforcement learning technology to simulate the operating data in the precise power supply management process to generate a new control strategy; The adjustment module continuously monitors the operating status of the power system in the cabin through the 5G network, collects actual operating data, combines the actual operating data with the new control strategy, adjusts the control strategy, and forms an intelligent power supply control solution.

9. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a cabin intelligent power supply control method based on 5G communication as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a method for controlling an intelligent power supply of a cabin based on 5G communication as described in any one of claims 1 to 7 is implemented.

Images