Battery replacement cabinet safety control method and system based on Internet of Things, and storage medium

Abstract

The invention relates to an internet-of-things-based power conversion cabinet safety control method and system and a storage medium, and the method comprises the steps: obtaining a predicted number of access times based on a preset prediction model according to a historical use record, and then obtaining an optimal control strategy based on the predicted number of access times according to temperature monitoring data and working condition data, thereby achieving the safety control of a power conversion cabinet. And finally, carrying out safety control on the target battery changing cabinet according to the optimal charging strategy. Compared with the prior art, on the control level, the safety of the battery replacement cabinet is improved by suspending charging of part of batteries, and the optimal control strategy is obtained by combining consideration of temperature and working conditions. In addition, the predicted number of times of use is predicted through a preset prediction model to serve as a constraint when the optimal control strategy is analyzed, and the practicability of the target power conversion cabinet is guaranteed. Most importantly, hardware is not improved, and the method has good universality. The problem of how to optimize the safety of the battery changing cabinet from the aspect of safety control strategies in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle power batteries, and in particular to a safety control method, system and storage medium for a battery swap cabinet based on the Internet of Things. Background Art

[0002] A battery swap cabinet is a facility that provides a quick battery replacement service for electric vehicles. Users can easily swap batteries without waiting for a lengthy charging process, significantly improving travel efficiency. The widespread use of battery swap cabinets is crucial for promoting the widespread use of electric vehicles and reducing urban carbon emissions.

[0003] However, for facilities like battery swap cabinets, since they retain a large number of batteries and are usually set up outdoors, there is a problem of long-term exposure to high temperature and sunlight. This requires the battery swap cabinet to have more stringent environmental adaptability and thermal management capabilities. Currently, the materials and construction of the battery swap cabinet must be able to withstand the impact of severe weather, including heat resistance, moisture resistance, and corrosion resistance. In addition, to meet the challenges brought by high temperatures, the battery swap cabinet should be equipped with an efficient thermal management system, such as a cooling fan, liquid cooling system, or phase change material, to keep the battery operating within a suitable temperature range and avoid performance degradation or safety risks due to overheating.

[0004] However, as mentioned above, existing battery swap cabinet safety technologies mostly focus on hardware structure improvements, which are costly to implement, while specific safety control aspects are still under research. Therefore, a battery swap cabinet safety control method that can optimize safety control strategies is needed. Summary of the Invention

[0005] Therefore, the present invention provides a power exchange cabinet safety control method, system and storage medium based on the Internet of Things to solve the problem of how to optimize the safety of the power exchange cabinet from the perspective of safety control strategy in the prior art.

[0006] The present invention provides a safety control method for a power exchange cabinet based on the Internet of Things, comprising:

[0007] Obtain the historical usage records of the target battery swap cabinet;

[0008] Based on historical usage records, the target battery swap cabinet is expected to be used for the next preset time period based on the preset prediction model;

[0009] Obtain temperature monitoring data and operating condition data of the target battery swap cabinet;

[0010] Based on the temperature monitoring data and operating condition data, and the expected number of times of use, the optimal control strategy is obtained. The optimal control strategy includes suspending the charging of some batteries in the target battery swap cabinet. The number of batteries whose charging is suspended is proportional to the temperature.

[0011] The target battery swap cabinet is safely controlled according to the optimal charging strategy.

[0012] The present invention also provides a preferred embodiment: the preset prediction model is a neural network model; based on the historical usage records, the expected number of times the target battery swap cabinet will be used in the next preset time period is predicted based on the preset prediction model, including:

[0013] Obtaining input data based on historical usage records, the input data including historical usage times and environmental data within a plurality of preset historical time periods in the past, wherein a total time span of the plurality of preset historical time periods in the input data exceeds six days;

[0014] An input vector is established based on the input data, and the input vector is input into a preset prediction model to obtain the expected number of times output by the preset prediction model.

[0015] The present invention also provides a preferred embodiment: the input vector includes a first vector, a second vector sequence and a third vector sequence, wherein each element in the first vector corresponds to a preset historical time period, and each element value represents the historical number of accesses within the preset historical time period; the second vector sequence includes multiple second vectors, each second vector corresponds to a preset historical time period, and the multiple element values ​​of each second vector respectively represent a variety of environmental data within the preset historical time period; the third vector sequence includes multiple third vectors, each third vector corresponds to a preset historical time period, and the multiple element values ​​of each third vector respectively represent the historical number of accesses and a variety of environmental data within the preset historical time period; the preset prediction model includes a first neural network, a second neural network, a third neural network and a fourth neural network, the first neural network is used to input the first vector to analyze the changing pattern of the historical number of accesses, the second neural network is used to input the second vector sequence to analyze the changing pattern of the environmental data, the third neural network is used to input the third vector sequence to analyze the degree of correlation between the historical number of accesses and the environmental data, the input end of the fourth neural network is simultaneously connected to the output ends of the first neural network, the second neural network and the third neural network, and the fourth neural network is used to output the expected number of accesses.

[0016] The present invention also provides a preferred embodiment: the first neural network, the second neural network and the third neural network are all recurrent neural networks, the first neural network outputs a first context vector, the second neural network outputs a second context vector, the third neural network outputs an associated feature value, the first context vector, the second context vector and the associated feature value are spliced ​​to form an intermediate vector and input into the fourth neural network.

[0017] The present invention also provides a preferred embodiment: according to the temperature monitoring data and the working condition data, based on the expected number of times of use, an optimal control strategy is obtained, including:

[0018] According to the working condition data and the expected number of times of use, multiple strategy description vectors are established, and the strategy description vector is used to represent a control strategy;

[0019] Based on temperature monitoring data and operating condition data, based on a preset fitness function, multiple strategy description vectors are optimized through a preset optimization algorithm to obtain the optimal control strategy.

[0020] The present invention also provides a preferred embodiment: the operating condition data includes the number of fully charged batteries, the number of charging batteries, the temperature of each charging battery and the remaining charging time of each charging battery, and the temperature monitoring data includes the temperature of each charging battery.

[0021] The present invention also provides a preferred embodiment: in the strategy description vector, each element corresponds to a charging battery, each element value represents the remaining charging time of the charging battery, and the time length represented by each element value is less than a preset time period.

[0022] The present invention also provides a preferred embodiment: the preset fitness function is:

[0023]

[0024] Wherein, f is the fitness of a strategy description vector, max is the preset upper threshold, x is an element value in the strategy description vector, T is the temperature of the charging battery corresponding to the element value, t is the remaining charging time of the charging battery corresponding to the element value, and a and b are different preset unit adjustment vectors.

[0025] The present invention also provides a power exchange cabinet safety control system based on the Internet of Things, comprising:

[0026] Data calling module, used to obtain the historical usage records of the target battery swap cabinet;

[0027] The usage analysis module is used to predict the expected number of times the target battery swap cabinet will be used in the next preset time period based on historical usage records and a preset prediction model;

[0028] Data acquisition module, used to obtain temperature monitoring data and operating condition data of the target battery swap cabinet;

[0029] The safety analysis module is used to obtain the optimal control strategy based on the expected number of accesses and temperature monitoring data and operating condition data. The optimal control strategy includes suspending the charging of some batteries in the target battery swap cabinet, and the number of batteries suspended is proportional to the temperature;

[0030] The safety control module is used to perform safety control on the target battery swap cabinet according to the optimal charging strategy.

[0031] The present invention also provides a computer-readable storage medium for storing computer-readable programs or instructions. When the program or instructions are executed by a processor, the steps in any of the above-mentioned methods for controlling the safety of power exchange cabinets based on the Internet of Things can be implemented.

[0032] The beneficial effects of adopting the above embodiment are:

[0033] The present invention provides a battery swap cabinet safety control method, system and storage medium based on the Internet of Things. The method first obtains the historical usage record of the target battery swap cabinet, and then predicts the expected number of times the target battery swap cabinet will be used in the next preset time period based on the historical usage record and a preset prediction model. Then, the temperature monitoring data and operating condition data of the target battery swap cabinet are obtained. Then, based on the temperature monitoring data and operating condition data and the expected number of times of use, an optimal control strategy is obtained. The optimal control strategy includes suspending the charging of some batteries in the target battery swap cabinet, and the number of batteries suspended from charging is proportional to the temperature. Finally, the target battery swap cabinet is safely controlled according to the optimal charging strategy. Compared with the prior art, the present invention improves the safety of the battery swap cabinet at the non-hardware control level by suspending the charging of some batteries. It obtains the optimal control strategy by combining temperature and operating condition considerations, so that the higher the temperature, the lower the number of batteries being charged, thereby improving the safety level. In addition, the present invention also predicts the expected number of uses through a preset prediction model, and uses the expected number of uses as a constraint when analyzing the optimal control strategy, thereby ensuring the practicality of the target battery swap cabinet and avoiding the problem that it cannot complete the basic battery supply function in pursuit of safety. Most importantly, the present invention is not a hardware improvement, so its implementation is not limited to specific equipment models and has good universal applicability. It solves the problem of how to optimize the safety of power exchange cabinets from the perspective of safety control strategy in the existing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A method flow chart of an embodiment of a power exchange cabinet safety control method based on the Internet of Things provided by the present invention;

[0035] Figure 2 A schematic diagram of the structure of a preset prediction model in one embodiment of the present invention;

[0036] Figure 3 This is a system structure diagram of an embodiment of the Internet of Things-based power exchange cabinet safety control system provided by the present invention. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0038] Combine Figure 1 As shown, a specific embodiment of the present invention discloses a safety control method for a power exchange cabinet based on the Internet of Things, including:

[0039] S101. Obtain historical usage records of the target battery swap cabinet;

[0040] S102: Based on historical usage records and a preset prediction model, predict the expected number of times the target battery swap cabinet will be used in the next preset time period;

[0041] S103, obtaining temperature monitoring data and operating condition data of the target battery swap cabinet;

[0042] S104. Based on the temperature monitoring data and operating condition data and the expected number of access times, an optimal control strategy is obtained. The optimal control strategy includes suspending charging of some batteries in the target battery swap cabinet, and the number of batteries whose charging is suspended is proportional to the temperature;

[0043] S105. Perform safety control on the target battery swap cabinet according to the optimal charging strategy.

[0044] Compared with the existing technology, the present invention improves the safety of the battery swap cabinet at the non-hardware control level by suspending the charging of some batteries. It obtains the optimal control strategy by combining the consideration of temperature and working conditions, so that when the temperature is higher, the number of batteries being charged is lower, thereby improving the safety level. In addition, the present invention also predicts the expected number of uses through a preset prediction model, and uses the expected number of uses as a constraint when analyzing the optimal control strategy, to ensure the practicality of the target battery swap cabinet and avoid the problem of being unable to complete the basic battery supply function in pursuit of safety. Most importantly, the present invention is not an improvement to the hardware, so that its implementation is not restricted by the specific equipment model and has good universality. It solves the problem of how to optimize the safety of the battery swap cabinet from the aspect of safety control strategy in the existing technology.

[0045] In step S102 of the above process, you can directly choose to perform statistical feature analysis on the historical usage records to obtain feature values ​​such as the average value, and then perform a simple linear transformation (for example, multiply the average value by a coefficient such as 1.2 as a retention margin) to obtain another value, which is used as the expected number of times used.

[0046] Furthermore, in a preferred embodiment, a more scientific and reasonable method for predicting the expected number of times the battery is used is provided. In this embodiment, the preset prediction model is a neural network model; the above step S102, based on the historical usage records, predicts the expected number of times the target battery swap cabinet will be used in the next preset time period based on the preset prediction model, including:

[0047] Obtaining input data based on historical usage records, the input data including historical usage times and environmental data within a plurality of preset historical time periods in the past, wherein a total time span of the plurality of preset historical time periods in the input data exceeds six days;

[0048] An input vector is established based on the input data, and the input vector is input into a preset prediction model to obtain the expected number of times output by the preset prediction model.

[0049] In the above process, a neural network model is used as a preset prediction model to analyze historical usage records more scientifically and rationally, and obtain a more accurate estimated number of times of use. It is understandable that in practice, the frequency of use of the battery swap cabinet is actually closely related to people's work cycles. Therefore, this embodiment further limits the time span of the input data input to the preset prediction model, that is, it must be at least more than six days, so that the input data implicitly reflects the fluctuation pattern of people's work cycles. In this way, there is no need to encode clear time information in the input vector input to the neural network model. On the one hand, it improves the convenience of coding and the development and operation efficiency of this embodiment. On the other hand, it ensures that the preset prediction model can grasp the use characteristics in the historical usage records and obtain more accurate prediction results.

[0050] Furthermore, in a preferred embodiment, the input vector includes a first vector, a second vector sequence and a third vector sequence, wherein each element in the first vector corresponds to a preset historical time period, and each element value represents the historical number of accesses within the preset historical time period; the second vector sequence includes multiple second vectors, each second vector corresponds to a preset historical time period, and the multiple element values ​​of each second vector respectively represent a variety of environmental data within the preset historical time period; the third vector sequence includes multiple third vectors, each third vector corresponds to a preset historical time period, and the multiple element values ​​of each third vector respectively represent the historical number of accesses and a variety of environmental data within the preset historical time period; the preset prediction model includes a first neural network, a second neural network, a third neural network and a fourth neural network, the first neural network is used to input the first vector to analyze the changing pattern of the historical number of accesses, the second neural network is used to input the second vector sequence to analyze the changing pattern of the environmental data, the third neural network is used to input the third vector sequence to analyze the degree of correlation between the historical number of accesses and the environmental data, the input end of the fourth neural network is simultaneously connected to the output ends of the first neural network, the second neural network and the third neural network, and the fourth neural network is used to output the expected number of accesses.

[0051] In the above process, environmental data refers to data related to the environment of the target battery swap cabinet, such as ambient temperature, humidity, light intensity, air quality, etc. The third vector sequence includes both the historical usage times and environmental data. Therefore, in fact, the third vector sequence can be directly input into a recurrent neural network model (such as RNN, LSTM, etc.) to obtain the expected usage times. However, in practice, it is undeniable that not all battery swap behaviors are related to the environment. For example, compared with the battery swap cabinets in the commercial area for leisure and entertainment, the usage times of the battery swap cabinets in the residential area are obviously more regular. Therefore, this embodiment has made improvements on conventional means, further dividing the input vector into a first vector, a second vector sequence and a third vector sequence, and further dividing the preset prediction model from a complete neural network into a first neural network, a second neural network, a third neural network and a fourth neural network. The first neural network and the second neural network are respectively analyzed based on the changing rules of the historical usage times and the environmental data themselves, while the third neural network is used to analyze the dependency relationship between the historical usage times and the environmental data, and then the fourth neural network is used to integrate the previous analysis results to obtain a more accurate prediction result.

[0052] In practice, the first neural network can only output a predicted number of usage times, the second neural network can only output a predicted environmental data vector, and the third neural network can only output a predicted value representing the relationship between the historical number of usage times and environmental data. In this way, the output of the first neural network can be corrected by the fourth neural network.

[0053] In a preferred embodiment, a more preferred method for connecting neural networks is provided. The first neural network, the second neural network, and the third neural network are all recurrent neural networks. The first neural network outputs a first context vector, the second neural network outputs a second context vector, and the third neural network outputs an associated feature value. The first context vector, the second context vector, and the associated feature value are concatenated to form an intermediate vector, which is then input into a fourth neural network.

[0054] Compared to the conventional scheme described above, the present invention limits the first neural network and the second neural network to recurrent neural networks, so that they can also output data in vector form, namely the first context vector and the second context vector. In this way, in addition to the predicted data, the output of the first neural network and the second neural network can actually be regarded as containing time-varying information (similar to the seq2seq model in natural language processing). In addition to the associated feature values ​​output by the third neural network, the fourth neural network can simultaneously analyze a large amount of meaningful data (far more than the three input data described above), and the fourth neural network can obtain more scientific and reasonable prediction results without using a recurrent neural network. Figure 2 This is a brief structural diagram of the preset prediction model in this embodiment.

[0055] Furthermore, in a preferred embodiment, the above step S104, obtaining the optimal control strategy based on the temperature monitoring data and the operating condition data and the expected number of times of use, includes:

[0056] According to the working condition data and the expected number of times of use, multiple strategy description vectors are established, and the strategy description vector is used to represent a control strategy;

[0057] Based on temperature monitoring data and operating condition data, based on a preset fitness function, multiple strategy description vectors are optimized through a preset optimization algorithm to obtain the optimal control strategy.

[0058] The main idea of ​​the present invention is to ensure safety by pausing the charging of some batteries according to the temperature. Therefore, in practice, after determining the expected number of times of use, the number of charged batteries to be retained can be determined according to the charging status of the target battery swap cabinet using any rule (for example, limiting the combined number of fully charged batteries and batteries that will be fully charged within a preset time period to be greater than 1-2 times the expected number of times of use), and making appropriate modifications according to the temperature (for example, the higher the temperature, the more appropriate it is to further reduce the number of batteries being charged, provided that the available batteries exceed the expected number of times of use) to ensure that the battery swap cabinet can be used normally. However, this method is obviously too rigid and not smart enough. In practice, battery charging is a dynamic process. The battery may be fully charged at any time within the preset time period, and the number of batteries being charged is not linearly related to the expected number of times of use. For example, retaining a certain number of batteries being charged with a longer charging time and retaining the same number of batteries being charged with a shorter charging time will have completely different actual user experiences. Therefore, this embodiment uses an optimization algorithm to solve this complex problem.

[0059] It is understandable that the optimization algorithm of this embodiment can adopt any existing optimization algorithm, such as genetic algorithm, PSO algorithm, etc., which will not be described in detail in this embodiment. However, it is understandable that the key to the application of these algorithms is the encoding method of the object to be optimized (i.e., the strategy optimization vector) and the determination of the fitness function. For example, for the PSO algorithm, the strategy optimization vector can be designed using a unique hot encoding: that is, each element in the vector represents a battery of a charging type, and the element value is 1 or 0, 1 means pausing its charging, and 0 means not pausing.

[0060] The present invention further provides a preferred design method: in a preferred embodiment, the operating condition data includes the number of fully charged batteries, the number of charging batteries, the temperature of each charging battery, and the remaining charging time of each charging battery (specifically, the time required for the battery to be fully charged), and the temperature monitoring data includes the temperature of each charging battery. In the strategy description vector, each element corresponds to a charging battery, and each element value represents the remaining charging time of the charging battery (different from the remaining charging time in the previous text, the remaining charging time here refers to the length of time the battery maintains the charging state before charging is suspended. Obviously, the remaining charging time of the same battery should be less than the remaining charging time). The length of time represented by each element value is less than the preset time period.

[0061] In this way, the strategy optimization vector in this embodiment not only includes the control strategy for whether to suspend battery charging, but also further includes the charging control strategy for the time span, which can obtain more precise and accurate results. For example, in the above design, the strategy description vector [3,1,6,0] can be interpreted as: suspend charging after 3 hours for battery 1, 1 hour for battery 2, 6 hours for battery 3, and directly suspend charging for battery 4. This method obviously obtains more practical control results than the modified method described above, and it can be achieved without making any improvements to the existing optimization algorithm in any steps other than the fitness function. In addition, during optimization, it is also necessary to constrain the legitimacy of the strategy description vector based on the expected number of accesses to avoid the phenomenon of insufficient available batteries. Establishing constraints is also a common method in optimization algorithms, so it will not be explained in detail in this article.

[0062] Furthermore, in a preferred embodiment, the preset fitness function is:

[0063]

[0064] Wherein, f is the fitness of a strategy description vector, max is the preset upper threshold, x is an element value in the strategy description vector, T is the temperature of the charging battery corresponding to the element value, t is the remaining charging time of the charging battery corresponding to the element value, and a and b are different preset unit adjustment vectors.

[0065] The significance of the above formula is to simultaneously evaluate the pros and cons of a control strategy represented by a strategy description vector from two dimensions: temperature and remaining charging time. If the remaining charging time (i.e., the element value of the strategy description vector) of a charging battery is shorter and the temperature is lower, its charging state should be retained. At this time, the fitness of the strategy description vector corresponding to the scheme is higher. The calculation result is guaranteed to be a positive number by presetting the upper and lower thresholds. At the same time, if the difference between the remaining charging time and the remaining charging time (i.e., the element value of the strategy description vector) of a charging battery is smaller, with 0 being the optimal value, its charging state should be retained. At this time, the fitness of the strategy description vector corresponding to the scheme is higher. In this way, the optimal charging strategy obtained is the most accurate and reasonable.

[0066] Combine Figure 3 As shown, the present invention also provides a power exchange cabinet safety control system based on the Internet of Things, including:

[0067] The data calling module 310 is used to obtain the historical usage records of the target battery swap cabinet;

[0068] The usage analysis module 320 is used to predict the expected number of times the target battery swap cabinet will be used in the next preset time period based on historical usage records and a preset prediction model;

[0069] The data acquisition module 330 is used to obtain the temperature monitoring data and operating condition data of the target battery swap cabinet;

[0070] Safety analysis module 340, used to obtain an optimal control strategy based on the expected number of accesses and temperature monitoring data and operating condition data. The optimal control strategy includes suspending charging of some batteries in the target battery swap cabinet, and the number of batteries suspended is proportional to the temperature;

[0071] The safety control module 350 is used to perform safety control on the target battery swap cabinet according to the optimal charging strategy.

[0072] It should be noted here that the corresponding system provided in the above embodiments can implement the technical solutions described in the above method embodiments. The specific implementation principles of the above modules or units can be found in the corresponding contents in the above method embodiments, which will not be repeated here.

[0073] This embodiment also provides a computer-readable storage medium on which a power exchange cabinet safety control program based on the Internet of Things is stored. When the power exchange cabinet safety control program based on the Internet of Things is executed by a processor, the steps in the above embodiments can be implemented.

[0074] The present invention provides a battery swap cabinet safety control method, system and storage medium based on the Internet of Things. The method first obtains the historical usage record of the target battery swap cabinet, and then predicts the expected number of times the target battery swap cabinet will be used in the next preset time period based on the historical usage record and a preset prediction model. Then, the temperature monitoring data and operating condition data of the target battery swap cabinet are obtained. Then, based on the temperature monitoring data and operating condition data and the expected number of times of use, an optimal control strategy is obtained. The optimal control strategy includes suspending the charging of some batteries in the target battery swap cabinet, and the number of batteries suspended from charging is proportional to the temperature. Finally, the target battery swap cabinet is safely controlled according to the optimal charging strategy. Compared with the prior art, the present invention improves the safety of the battery swap cabinet at the non-hardware control level by suspending the charging of some batteries. It obtains the optimal control strategy by combining temperature and operating condition considerations, so that the higher the temperature, the lower the number of batteries being charged, thereby improving the safety level. In addition, the present invention also predicts the expected number of uses through a preset prediction model, and uses the expected number of uses as a constraint when analyzing the optimal control strategy, thereby ensuring the practicality of the target battery swap cabinet and avoiding the problem that it cannot complete the basic battery supply function in pursuit of safety. Most importantly, the present invention is not a hardware improvement, so its implementation is not limited to specific equipment models and has good universal applicability. It solves the problem of how to optimize the safety of power exchange cabinets from the perspective of safety control strategy in the existing technology.

[0075] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referenced to each other.

[0076] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A safety control method for a power exchange cabinet based on the Internet of Things, characterized in that: include: Obtain the historical usage records of the target battery swap cabinet; According to the historical usage records, the target battery swap cabinet is expected to be used in the next preset time period based on the preset prediction model; Obtain temperature monitoring data and operating condition data of the target power swap cabinet; According to the temperature monitoring data and working condition data, based on the expected number of times of use, the optimal control strategy is obtained. The optimal control strategy includes suspending the charging of some batteries in the target battery swap cabinet. The number of batteries suspended from charging is proportional to the temperature. The target battery swap cabinet is safely controlled according to the optimal charging strategy.

2. The method for controlling the power exchange cabinet based on the Internet of Things according to claim 1 is characterized in that: The preset prediction model is a neural network model; based on the historical usage records, the preset prediction model predicts the expected number of times the target battery swap cabinet will be used in the next preset time period, including: According to the historical usage records, input data is obtained, the input data including the historical usage times and environmental data in a plurality of preset historical time periods in the past, wherein the total time span of the plurality of preset historical time periods in the input data exceeds six days; According to the input data, an input vector is established, and the input vector is input into a preset prediction model to obtain the estimated number of times output by the preset prediction model.

3. The method for controlling the power exchange cabinet based on the Internet of Things according to claim 2 is characterized in that: The input vector includes a first vector, a second vector sequence and a third vector sequence, wherein each element in the first vector corresponds to a preset historical time period, and each element value represents the historical number of times of use within the preset historical time period; the second vector sequence includes multiple second vectors, each second vector corresponds to a preset historical time period, and the multiple element values ​​of each second vector respectively represent a variety of environmental data within the preset historical time period; the third vector sequence includes multiple third vectors, each third vector corresponds to a preset historical time period, and the multiple element values ​​of each third vector respectively represent the historical number of times of use and a variety of environmental data within the preset historical time period; the preset prediction model includes a first neural network, a second neural network, a third neural network and a fourth neural network, the first neural network is used to input the first vector to analyze the changing law of the historical number of uses, the second neural network is used to input the second vector sequence to analyze the changing law of the environmental data, the third neural network is used to input the third vector sequence to analyze the correlation between the historical number of uses and the environmental data, the input end of the fourth neural network is simultaneously connected to the output ends of the first neural network, the second neural network and the third neural network, and the fourth neural network is used to output the expected number of uses.

4. The method for controlling the power exchange cabinet based on the Internet of Things according to claim 3 is characterized in that: The first neural network, the second neural network and the third neural network are all recurrent neural networks. The first neural network outputs a first context vector, the second neural network outputs a second context vector, and the third neural network outputs an associated feature value. The first context vector, the second context vector and the associated feature value are concatenated to form an intermediate vector and input into the fourth neural network.

5. The method for controlling power exchange cabinet safety based on the Internet of Things according to claim 1 is characterized in that: According to the temperature monitoring data and working condition data, based on the expected number of times of use, the optimal control strategy is obtained, including: According to the working condition data and based on the expected number of times of use, multiple strategy description vectors are established, and the strategy description vector is used to represent a control strategy; Based on temperature monitoring data and operating condition data, based on a preset fitness function, multiple strategy description vectors are optimized through a preset optimization algorithm to obtain the optimal control strategy.

6. The method for controlling the power exchange cabinet based on the Internet of Things according to claim 5 is characterized in that: The operating condition data includes the number of fully charged batteries, the number of charging batteries, the temperature of each charging battery, and the remaining charging time of each charging battery, and the temperature monitoring data includes the temperature of each charging battery.

7. The method for controlling the power exchange cabinet based on the Internet of Things according to claim 6 is characterized in that: In the strategy description vector, each element corresponds to a battery being charged, each element value represents the remaining charging time of the battery being charged, and the time length represented by each element value is less than a preset time period.

8. The method for controlling the power exchange cabinet based on the Internet of Things according to claim 6 is characterized in that: The default fitness function is: Among them, f is the fitness of a strategy description vector, max is the preset upper limit threshold, x is an element value in the strategy description vector, T is the temperature of the charging battery corresponding to the element value, t is the remaining charging time of the charging battery corresponding to the element value, and a and b are different preset unit adjustment vectors.

9. A power exchange cabinet safety control system based on the Internet of Things, characterized in that: include: A data calling module is used to obtain the historical usage records of the target battery swap cabinet; The usage analysis module is used to predict the expected number of times the target battery swap cabinet will be used in the next preset time period based on the historical usage records and the preset prediction model; A data acquisition module is used to obtain temperature monitoring data and operating condition data of the target power exchange cabinet; The safety analysis module is used to obtain the optimal control strategy based on the expected number of accesses according to the temperature monitoring data and working condition data. The optimal control strategy includes suspending the charging of some batteries in the target battery swap cabinet, and the number of batteries suspended from charging is proportional to the temperature; The safety control module is used to perform safety control on the target battery swap cabinet according to the optimal charging strategy.

10. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the Internet of Things-based power exchange cabinet safety control method described in any one of claims 1 to 8.

Images