Battery Management Device for Low- and High-Temperature Environments

Abstract

The present invention is a battery management device for responding to low and high temperature environments, and may include a management server including a temperature sensing sensor.

Description

Technology Field

[0001] The present invention relates to a battery management device for low and high temperature environments. Background Technology

[0002] Batteries are used as a key energy source in various application fields, such as electric vehicles, energy storage devices, and portable electronic devices.

[0003] However, battery performance varies significantly depending on the temperature of the operating environment; in particular, in low-temperature environments, the rate of internal chemical reactions slows down, causing a sharp drop in output and charging efficiency, while in high-temperature environments, safety issues such as electrolyte decomposition or thermal runaway may occur.

[0004] Therefore, operating the battery within a constant, appropriate temperature range is essential for maintaining performance, extending its lifespan, and ensuring safety.

[0005] In existing technologies, thermal control was achieved by simply attaching heaters or cooling devices to the outside of the battery; however, because this relied solely on surface temperature and could not directly reflect the internal temperature of the battery, precise control was difficult and unnecessary energy consumption occurred.

[0006] In addition, when managing multiple batteries simultaneously, there was a limitation in that efficient operation was restricted because it was difficult to distinguish and manage the status of individual batteries in real time.

[0007] To solve these problems, a temperature sensing technology capable of precisely estimating the internal temperature of the battery, and a management system that efficiently controls a heating module and a cooling module that perform heating at low temperatures and cooling at high temperatures based on this are required.

[0008] Meanwhile, the aforementioned background technology is technical information that the inventor possessed for the derivation of the present invention or acquired during the process of deriving the present invention, and it cannot necessarily be considered publicly known technology disclosed to the general public prior to the filing of the present invention. Prior art literature

[0009] Korean Published Patent No. 1020250056480 The problem to be solved

[0010] The objective of the present invention is to provide a battery management device capable of accurately estimating the internal temperature of a battery and maintaining the battery within an appropriate temperature range in response to both low-temperature and high-temperature environments.

[0011] The technical problems of the present invention are not limited to those mentioned above, and other unmentioned technical problems will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0012] A battery management device for low and high temperature environments according to one embodiment of the present invention may include a management server including a temperature sensing sensor.

[0013] According to one embodiment, the management server may include: a temperature sensing module that detects and estimates the internal temperature of at least one battery in real time; a heating control module that controls a heating module provided on one side of the battery to heat the battery when the temperature is low and release heat; a cooling control module that controls a cooling module provided on one side of the battery to cool the battery when the temperature is high and release cold air; and a temperature correction module that generates a control command to control the operation of the heating module and the cooling module so that a preset appropriate temperature range of the battery is maintained, and transmits the generated control command to the heating control module or the cooling control module.

[0014] According to one embodiment, the temperature sensing module estimates the internal temperature of the battery based on a plurality of temperature sensing sensors already provided on the outside of the battery and the surface temperature and ambient temperature of the battery detected by the plurality of temperature sensing sensors, and can estimate the internal temperature of the battery by reflecting the surface temperature and ambient temperature of the battery and model information of the battery.

[0015] According to one embodiment, the temperature sensing module communicates in real time with the temperature correction module through a network, assigns a unique ID to each battery and stores it in a database, and can simultaneously provide the unique ID of each battery, the current internal temperature of the battery, and the current time to the temperature correction module.

[0016] According to one embodiment, the temperature sensing module may provide the unique ID of each battery, the current internal temperature of the battery, and the current time to the temperature correction module at predetermined intervals.

[0017] According to one embodiment, the pre-set optimal temperature range of the battery may mean a range between the lowest temperature and the highest temperature input from a portable manager terminal while communicating with the management server.

[0018] According to one embodiment, the temperature correction module can generate a control command to operate the heating module and stop the operation of the cooling module when the current internal temperature of a specific battery is lower than a preset appropriate temperature range of the battery, and transmit the command to the heating control module and the cooling control module.

[0019] According to one embodiment, the temperature correction module can generate a control command to operate the cooling module and stop the operation of the heating module when the current internal temperature of a specific battery is higher than a preset appropriate temperature range of the battery, and transmit the command to the heating control module and the cooling control module.

[0020] According to one embodiment, the heating module may refer to a module capable of generating heat to raise the internal temperature of a battery, including a PTC heater, a heat pad, and a Peltier.

[0021] According to one embodiment, the heating module can continuously vary the amount of heat generated by a control command received from the heating control module.

[0022] According to one embodiment, the cooling module may refer to a module capable of releasing cold air to lower the internal temperature of the battery, including a fan, pump, valve, liquid cooling, heat sink, and Peltier.

[0023] According to one embodiment, the cooling module can continuously vary the amount of cooling by means of a control command received from the cooling control module.

[0024] According to one embodiment, if there is a battery that falls outside the error range of a preset appropriate temperature range of the battery, the temperature correction module can send the unique ID of the battery, the current internal temperature of the battery, and the current time to an administrator terminal.

[0025] According to one embodiment, the error range of the pre-set optimal temperature range of the battery may refer to a range pre-entered by the administrator terminal.

[0026] According to one embodiment, the temperature sensing module utilizes a logarithmic function that reduces the variance of the prediction value by reducing the influence of outliers and a trigonometric function that reduces the residual of the response by offsetting iterative errors to estimate the internal temperature (T in By utilizing a control decision algorithm that extracts (t), the internal temperature of the battery can be estimated by reflecting noise, outliers, and periodicity filtering.

[0027] According to one embodiment, the control judgment algorithm comprises the surface temperature (T) of the battery. sIt may include a first element that reflects the direct effect of ) on the internal temperature of the battery.

[0028] According to one embodiment, the first element is a surface temperature (T s It can be characterized by being configured to reflect the real-time temperature observed by a temperature sensing sensor that detects ) in the internal temperature estimation of the battery by weighting it.

[0029] According to one embodiment, the control judgment algorithm comprises the ambient temperature (T of the battery) a It may include a second element that reflects background influences, including external air inflow and heat dissipation paths, on the internal temperature of the battery.

[0030] According to one embodiment, the second element may be characterized by being configured to include a reality in which the external environment continuously affects the internal thermal balance.

[0031] According to one embodiment, the control judgment algorithm comprises a surface temperature (T s ) and ambient temperature (T a It may include a third factor that reflects the tendency for the influence of temperature transferred into the battery to increase as the difference of ) increases.

[0032] According to one embodiment, the third element is a surface temperature (T s ) and ambient temperature (T a A logarithmic function is utilized to suppress outlier runaways by correcting for a gradual increase even if the difference of ) widens, and the internal temperature estimate (T) is obtained by multiplying by a temperature scaling factor (α). in It can be characterized by adjusting the importance reflected in (t). Effects of the invention

[0033] According to one aspect of the present invention described above, the battery management device for low and high temperature environments proposed by the present invention can accurately estimate the internal temperature of the battery and maintain the battery in an appropriate temperature range in response to both low and high temperature environments.

[0034] The effects of the present invention are not limited to those mentioned above, and various effects may be included within the scope obvious to a person skilled in the art from the contents described below. Brief explanation of the drawing

[0035] FIG. 1 is a conceptual diagram of a battery management device for low and high temperature environments according to one embodiment of the present invention. FIG. 2 is a conceptual diagram of a management server according to one embodiment of the present invention. Figure 3 illustrates an overall relationship diagram according to the present invention. Figure 4 illustrates a flowchart between all components according to the present invention. Figure 5 illustrates a flowchart of process 1 according to the present invention. Figure 6 illustrates a flowchart of process 2 according to the present invention. Figure 7 illustrates a flowchart of process 3 according to the present invention. FIG. 8 illustrates a flowchart of process 4 according to the present invention. Specific details for implementing the invention

[0036] The following detailed description of the invention refers to the accompanying drawings, which illustrate specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. It should be understood that various embodiments of the invention are different but need not be mutually exclusive. For example, specific shapes, structures, and characteristics described herein may be implemented in other embodiments without departing from the spirit and scope of the invention in relation to one embodiment.

[0037] When it is stated that one component is "connected" or "contracted" to another component, it should be understood that while it may be directly connected or contracted to that other component, there may also be other components in between. Conversely, when it is stated that one component is "directly connected" or "directly contracted" to another component, it should be understood that there are no other components in between.

[0038] Furthermore, it should be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the invention. Accordingly, the following detailed description is not intended to be taken in a limiting sense, and the scope of the invention is limited only by the appended claims, including all equivalents thereof, provided appropriately described. Similar reference numerals in the drawings refer to the same or similar functions across various aspects.

[0039] Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the drawings.

[0041] FIG. 1 is a conceptual diagram of a battery management device for low and high temperature environments according to one embodiment of the present invention.

[0042] Referring to FIG. 1, a battery management device for low and high temperature environments according to one embodiment of the present invention may include a management server (100), a temperature sensing sensor (500), and a manager terminal (300).

[0043] The management server (100) can estimate the temperature of the battery by communicating with a temperature sensing sensor provided on the surface and surrounding of a plurality of batteries according to the present invention, and can individually recognize a plurality of batteries and generate and transmit an optimal control command in real time according to the state of each battery.

[0044] A temperature sensing sensor (500) is provided on the outside of the battery and may be composed of multiple sensors, and each sensor can detect the surface temperature of the battery and the ambient temperature and provide the detection information to the management server (100) through a network.

[0045] The administrator terminal (300) may refer to a terminal connected to a management server that can set and monitor the temperature control of the battery.

[0046] Additionally, the administrator terminal (300) can receive input from the administrator regarding an appropriate temperature range defined by the lowest and highest temperatures of the battery and provide it to the management server (100). When the internal temperature of a specific battery deviates from the preset range, the administrator terminal can receive the ID of the battery, the current internal temperature, and the current time from the management server (100) and output them to the display.

[0047] The management server (100) and the administrator terminal (300) may be a self-contained server or a cloud server for providing the service according to the present invention, or may be a peer-to-peer (P2P) set of distributed nodes.

[0048] The management server (100) can perform one or more of the operations, storage, reference, input / output, and control functions of a general computer, and may include an artificial neural network described later based on input data.

[0049] The management server (100) may include a processor and memory. The processor may include devices capable of individually recognizing a plurality of batteries according to the present invention, generating and transmitting optimal control commands in real time according to the state of each battery, and performing such actions. The processor may execute a program or control the management server (100). Program code executed by the processor may be stored in memory. The memory may store relevant information for performing a service according to the present invention or a program for implementing a method. The memory may be volatile memory or non-volatile memory.

[0050] The management server (100) can send data to an external device or receive data from an external device using a network.

[0051] The management server (100) can train an artificial neural network and can also use an artificial neural network that has been trained. The processor can train or execute an artificial neural network stored in memory, and the memory can store an artificial neural network that has been trained. The electronic device that trains the artificial neural network and the electronic device that uses it may be the same, but they may also be separate.

[0052] Artificial intelligence is a computer system that partially implements the functions of the human brain and is capable of learning, speculating, and making judgments on its own. As learning progresses, the probability of extracting the correct answer can increase. Artificial intelligence can be composed of learning and component technologies that utilize it. The learning aspect of AI is an algorithmic technology that classifies and learns features based on input data, while the component technologies may be techniques that utilize these learning algorithms to partially implement the functions of the human brain.

[0053] Artificial intelligence is a technology that facilitates the approach to problems where multiple probabilistic answers are possible, enabling it to logically and probabilistically infer optimal cycles, methods, and plans based on input data. AI inference techniques can include evaluating input data, optimization prediction, knowledge and probability-based reasoning, and preference-based planning.

[0054] Artificial neural networks are learning algorithms in the field of machine learning that programmatically implement the connections between neurons and synapses in the brain. By creating a neural network structure through programming and then training it, artificial neural networks can acquire desired functions. Although errors may exist, they can learn from massive datasets to produce appropriate output data from input data. They have the advantage of being able to obtain output data that has yielded statistically good results and are similar to human reasoning.

[0055] The management server (100) can infer individual characteristics and interests by analyzing consumers' online behavior data, social media activities, search history, etc., using an artificial intelligence algorithm built based on big data, and may include a number of pre-trained artificial neural networks for this purpose.

[0056] The network is a high-speed backbone network of a large-scale communication network capable of high-capacity, long-distance voice and data services, and may be a next-generation wired and wireless network for providing the Internet or high-speed multimedia services.

[0057] If the network is a mobile communication network, it may be a synchronous mobile communication network or an asynchronous mobile communication network. As an example of an asynchronous mobile communication network, a WCDMA (Wideband Code Division Multiple Access) network may be cited. In this case, although not shown in the drawing, the network may include an RNC (Radio Network Controller). Meanwhile, although a WCDMA network was given as an example, it may be a 3G LTE network, a 4G network, a next-generation communication network such as 5G, or other IP-based IP networks.

[0058] The management server (100) and the administrator terminal (300) may include any terminal capable of exchanging data over a network, such as a desktop computer, laptop, tablet, or smartphone.

[0059] The management server (100) and the administrator terminal (300) may include one or more of the computation function, storage function, reference function, input / output function, and control function of a computer to perform the service according to the present invention.

[0060] The management server (100) and the administrator terminal (300) may access a website or install an application to receive the service according to the present invention. The management server (100) and the administrator terminal (300) may exchange data through the website or the application.

[0061] The network is a high-speed backbone network of a large-scale communication network capable of high-capacity, long-distance voice and data services, and may be a next-generation wired and wireless network for providing the Internet or high-speed multimedia services.

[0062] If the network is a mobile communication network, it may be a synchronous mobile communication network or an asynchronous mobile communication network. As an example of an asynchronous mobile communication network, a WCDMA (Wideband Code Division Multiple Access) network may be cited. In this case, although not shown in the drawing, the network (300) may include an RNC (Radio Network Controller). Meanwhile, although a WCDMA network was given as an example, it may be a 3G LTE network, a 4G network, a 5G network, or other next-generation communication networks, or other IP-based IP networks.

[0063] A system (1) according to one embodiment of the present invention can individually recognize a plurality of batteries and generate and transmit optimal control commands in real time according to the state of each battery.

[0065] FIG. 2 is a conceptual diagram of a management server according to one embodiment of the present invention.

[0066] Referring to FIG. 2, a management server (100) according to one embodiment of the present invention may include a temperature sensing module (110), a heating control module (130), a cooling control module (150), and a temperature correction module (170).

[0067] The temperature sensing module (110) can detect and estimate the internal temperature of at least one battery in real time.

[0068] A temperature sensing module (110) can estimate the internal temperature of a battery based on a plurality of temperature sensing sensors (500) already provided on the outside of the battery and the surface temperature and ambient temperature of the battery detected by the plurality of temperature sensing sensors (500), and can estimate the internal temperature of the battery by reflecting the surface temperature and ambient temperature of the battery and the model information of the battery.

[0069] The temperature sensing module (110) communicates in real time with the temperature correction module (170) via a network, assigns a unique ID to each battery and stores it in a database, and can simultaneously provide the unique ID of each battery, the current internal temperature of the battery, and the current time to the temperature correction module (170).

[0070] The temperature sensing module (110) can provide the unique ID of each battery, the current internal temperature of the battery, and the current time to the temperature correction module (170) at predetermined intervals.

[0071] The optimal temperature range of the battery set in advance communicates with the management server (100) and may refer to a range between the lowest temperature and the highest temperature input from a portable manager terminal (300).

[0072] Meanwhile, the temperature sensing module (110) may utilize a control judgment algorithm as a means of correction and normalization to simultaneously achieve stability and realism for estimating the internal temperature of the battery.

[0073] If only the surface or ambient temperature is used alone, it may be difficult to represent the internal thermal state of the battery, so the temperature sensing module (110) may use a control judgment algorithm to calculate an estimated value that filters and reflects noise, outliers, and periodicity.

[0074] The control judgment algorithm can be utilized in the internal temperature estimate extraction step of the temperature sensing module (110), and the temperature sensing module (110) uses the internal temperature estimate (T) extracted by the control judgment algorithm. in (t)) can be provided as a temperature correction module (170) to enable the generation of driving judgment and control commands for the heating module (700) or cooling module (900) in a subsequent step.

[0075] In the process described above, correction using a logarithmic function utilized in the control decision algorithm can reduce the influence of outliers and decrease the variance of the predictions, and periodic correction can offset iterative errors to mitigate the residuals of the response.

[0076] The control decision algorithm is the battery surface temperature (T s It may include a first element that reflects the direct influence of ) on the internal temperature of the battery, which is the surface temperature (T s It can be characterized by being configured to reflect the real-time temperature observed by a temperature sensing sensor that detects ) in the estimation of the battery's internal temperature by weighting it.

[0077] In addition, the control decision algorithm is the battery's ambient temperature (T a It may include a second element that reflects the background influence (external air inflow, heat dissipation path, etc.) on the internal temperature of the battery, and may be characterized by being configured to include the reality that the external environment continuously affects the internal thermal balance.

[0078] In addition, the control decision algorithm is the surface temperature (T s ) and ambient temperature (T a It may include a third factor that reflects the tendency for the influence of temperature transferred into the battery to increase as the difference of ) increases.

[0079] The third element is the surface temperature (T) by using a logarithmic function. s ) and ambient temperature (T a Even if the difference of ) increases, it can be corrected to increase gradually to suppress the runaway of outliers, and by multiplying by the temperature scale factor (α), the internal temperature estimate (T in The importance reflected in (t) can be adjusted.

[0080] In addition, the control decision algorithm may include a fourth element that reflects errors caused by time periodicity (load cycles, fan or pump cycles, etc.) that commonly occur in battery operating environments, in order to offset them with a small amplitude.

[0081] The fourth element may be characterized by being configured to correct the value obtained by dividing time (t) by the time scale coefficient (τ) by the input of the trigonometric function and then multiplying it by the amplitude coefficient (β), which is the amplitude unit.

[0082] The control decision algorithm directly or immediately affects the surface temperature (T) of the battery's internal temperature. s Using ) the internal temperature estimate (T in (t)) can be extracted.

[0083] Surface temperature (T s ) may refer to a temperature value obtained by sampling at intervals of 1 to 5 seconds through a temperature sensing sensor consisting of one of a thermistor, RTD, or thermocouple attached to the outer surface of the battery, and may use °C units.

[0084] The control decision algorithm reflects the external environmental influence of the external thermal environment on internal thermal equilibrium and estimates the internal temperature (T in To extract (t)), the ambient temperature (T a You can utilize ).

[0085] Ambient temperature (T a ) may refer to a temperature value obtained by sampling at intervals of 1 to 5 seconds through an ambient temperature sensing sensor placed in the external air layer of the battery pack or in the intake path and exhaust path, and may use °C units.

[0086] Ambient temperature (T a A temperature sensing sensor installed to collect ) may be provided in a space where the turbulent or radiative effects of the space where the battery is located are not excessive.

[0087] The control decision algorithm expresses the periodic patterns associated with the operation and control of the battery within the algorithm and the internal temperature estimate (T in Time (t) can be utilized to reflect in the extraction of (t).

[0088] Time (t) may mean the time elapsed since the battery started operating, that is, the time elapsed since the measurement session started (time elapsed from the time the measurement session started (e.g., the time when the battery power was applied, the time when the management server first registered the battery-ID, etc.)) and may be obtained from a timestamp in the system clock of the management server (100) and may be used in units of seconds (s).

[0089] The control decision algorithm determines the battery surface temperature and ambient temperature based on the internal temperature estimate (T in (t)) Surface temperature weight (w1) and ambient temperature weight (w2) can be utilized to adjust the relative ratio of the influence on the extraction.

[0090] The surface temperature weight (w1) and ambient temperature weight (w2) are each the surface temperature (T s ) and ambient temperature (T a As a dimensionless coefficient representing the contribution of ), the initial value can be set within the range of 0.6 to 0.8 and 0.2 to 0.4, respectively.

[0091] After this, the surface temperature weight (w1) and ambient temperature weight (w2) are the collected surface temperature (T s ) and ambient temperature (T a ) and, surface temperature (T s ) and ambient temperature (T a The internal temperature estimate (T) extracted from ) in (t)) learns the surface temperature or ambient temperature as an internal temperature estimate (T in (t)) The ratio of influence on extraction can be adjusted.

[0092] The control decision algorithm is the surface temperature (T s ) and ambient temperature (Ta To reflect the additional upward or downward trend caused by the difference in ) within the battery without excessively, the internal temperature estimate (T) is calculated using a temperature scaling factor (α). in (t)) can be extracted.

[0093] The temperature scale factor (α) is a factor that determines the size of the third element corrected by a logarithmic function, and can use °C units. As a default value, the initial setting value can be configured by querying the model-specific recommended range from the battery model DB (cell type, pack structure, thermal resistance / thermal capacity metadata).

[0094] The control decision algorithm matches the units of the fourth element and utilizes the amplitude coefficient (β) to limit the correction size for the internal temperature estimate (T in (t)) can be extracted.

[0095] The amplitude factor (β) can be a factor for controlling the amplitude of the fourth element, and can be expressed in °C units. It can be derived by estimating the RMS magnitude of the high-frequency residual of the internal temperature in the battery's operating log and reflecting a portion of it (e.g., 30–70%).

[0096] The control decision algorithm utilizes the time scale factor (τ) to determine the speed and suitability of the period correction of time (t) used in the fourth element, and the internal temperature estimate (T in (t)) can be extracted.

[0097] The time scale factor (τ) can refer to a value set near the strongest peak period by detecting the periodicity of the operation log (fan PWM, pump duty, charge / discharge cycle) through autocorrelation or frequency analysis.

[0098] [Table 1]

[0099]

[0100] As shown in [Table 1] above, the surface temperature (T s ) and ambient temperature (Ta It can be observed that even if the difference of ) changes significantly, it increases gradually when corrected with a logarithmic function, and through this, the control decision algorithm can be characterized by using a logarithmic function to suppress the influence of outliers.

[0101] In addition, as shown in [Table 1] above, it can be seen that when time (t) is corrected together with trigonometric functions, the periodicity of time is smoothly offset, and through this, the control decision algorithm can be characterized by using trigonometric functions to smoothly reflect the time variable.

[0102] As shown in [Table 1] and the description above, the location recognition reliability algorithm may be characterized by being designed using functions such as logarithmic functions through examples of various tests, and may be characterized by being designed to reflect actual data distribution and experimental experience.

[0103] As shown in [Table 1] and the above description, the control decision algorithm may be characterized by being constructed using the aforementioned variables and functions to solve field problems such as lack of internal temperature representativeness, sensitivity to outliers, and periodic residuals.

[0104] First, the control decision algorithm can obtain a basic representative value as a weighted sum of surface temperature and ambient temperature, which allows for easy adjustment of sensor reliability differences and thermal path differences using two weights.

[0105] Next, the control decision algorithm models the empirical fact that the tendency for heat inflow or outflow to increase as the temperature difference increases using a logarithmic function, thereby ensuring the gradualness of the correction amount even under rapid changes in the external environment.

[0106] Finally, the control decision algorithm can reduce fluctuations in the estimate by offsetting iterative residuals caused by device operating characteristics, such as fans, pumps, and load cycles, using trigonometric functions.

[0107] For example, surface temperature (T s ) is 12°C, ambient temperature (T a When ) is derived as 5°C and time (t) as 1800s, and the surface temperature weight (w1) is set to 0.7, the ambient temperature weight (w2) to 0.2, the temperature scale factor (α) to 1.5°C, the amplitude factor (β) to 0.5°C, and the time scale factor (τ) to 900s, the internal temperature estimate (T in (t) can be extracted as approximately 12.97.

[0108] The internal temperature estimate (T) extracted by the example described above in (t)) is provided to the temperature correction module (170) and compared with the appropriate temperature range and can be used as a direction for the control command of the heating module (700) or cooling module (900).

[0109] For example, the internal temperature estimate (T in If (t)) is extracted at a value lower than a threshold value pre-set by the user, a control command to drive the heating module (700) can be generated by the temperature correction module (170).

[0110] The control decision algorithm can be implemented in the collection phase, preprocessing phase, extraction phase, decision and transmission phase, and storage and monitoring phase.

[0111] In the collection stage of the control judgment algorithm, the temperature sensing module (110) detects the surface temperature (T) at predetermined intervals (e.g., 1 second, 5 seconds, etc.). s ), ambient temperature (T a ), time (t) can be obtained from the timestamps of each temperature sensing sensor and the management server.

[0112] In the preprocessing step of the control judgment algorithm, the temperature sensing module (110) can apply a time scale factor (τ) to the acquired time and remove outliers.

[0113] In the extraction step of the control judgment algorithm, the temperature sensing module (110) applies each acquired variable to the preprocessed value to obtain an internal temperature estimate (T in (t)) can be extracted.

[0114] In the judgment and transmission step of the control judgment algorithm, the temperature sensing module (110) obtains the extracted internal temperature estimate (T in (t)) can be transmitted to the temperature correction module (170) to support the determination of the heating module (700) or cooling module control command.

[0115] In the storage and monitoring stage of the control judgment algorithm, the temperature sensing module (110) (battery unique number (ID), internal temperature estimate (T) in It can be stored in a database in the form of (t)), time (t)) and monitored in real-time status.

[0116] The control decision algorithm determines the appropriate temperature range (e.g., T) from the server administrator. min =5°C, T max It can receive and store (=35°C), and the specific range may vary depending on external environmental conditions.

[0117] Internal temperature estimate (T) extracted from the control decision algorithm in (t)) is T min If extracted to be smaller, it can be determined that the operating conditions of the heating module (700) are satisfied, and the internal temperature estimate (T in (t)) is T max If a larger amount is extracted, it can be determined that the operating conditions of the cooling module (900) are satisfied.

[0118] The operation of the heating module (700) and the cooling module (900) can be designed to be mutually exclusive so as not to be operated simultaneously, and if necessary, can be extended to a weighted combination method such as alarm score calculation, but in the control judgment algorithm, a simple threshold-based exclusive logic may be preferred to ensure clarity of control.

[0119] Regarding the processing logic and condition changes according to the time flow of the control judgment algorithm, the temperature sensing module (110) derives each variable with periodic sampling of 1 to 5 seconds and repeatedly calculates the process of inputting it into a function to obtain an internal temperature estimate (T in A time series of (t) can be generated, and by constantly monitoring the average and change of the most recent N samples, an internal temperature estimate (T in (t)) may also detect abrupt changes that exceed a predetermined threshold.

[0120] Considering that the battery's operating pattern may change, the control decision algorithm can maintain the period correction in alignment with the actual operating cycle by re-entering or re-deriving the time scale factor (τ) daily or periodically.

[0121] In addition, the control decision algorithm can prevent over-correction or under-correction by slightly adjusting the amplitude coefficient (β), which is the correction amplitude constant, when the level of environmental noise changes, and this time-based processing logic can simultaneously achieve performance that responds to long-term trend changes while suppressing short-term noise.

[0122] The control judgment algorithm is determined by the server administrator through the management server (100) or the temperature sensing module (110) to determine an appropriate temperature range (e.g., T min =5°C, T max By adjusting the surface temperature (T = 35°C) to the same input value, the same input value surface temperature (T s ), ambient temperature (T a The result of operating the heating module or cooling module may also vary depending on the time (t).

[0123] Additionally, the control judgment algorithm can adjust the contribution ratio of the surface temperature and ambient temperature by receiving initial values ​​of the surface temperature weight (w1) and ambient temperature weight (w2) from the server administrator through the management server (100) or the temperature detection module (110) in accordance with external environmental conditions, or by performing a simple calibration.

[0124] Additionally, the amplitude coefficient (β) can be adjusted to change the sensitivity of periodicity reflection, and the time scale coefficient (τ) can be updated to improve the synchronization of periodic correction when the battery operation schedule or fan control strategy changes, and such user input consequently the internal temperature estimate (T in It can have a direct impact on the interpretation of (t) and control decision-making, including heating and cooling modules.

[0125] Each variable and item of the control decision algorithm is specifically defined, measured, and configured to be summable without unit conflicts, thereby providing reliable results and achieving technical effects such as improved precision and effectiveness compared to existing methods, which can be clearly understood and implemented even by ordinary technicians.

[0126] Meanwhile, the temperature sensing module (110) may utilize a temperature sensor judgment algorithm as a quantitative judgment means to quickly determine whether the temperature sensing sensor is operating correctly.

[0127] The temperature sensor determination algorithm is utilized in real time by the temperature detection module (110), and the extracted result can be provided simultaneously to the temperature correction module (170) and the database, and can also be used as a trigger to send an alarm to the administrator terminal (300) when necessary.

[0128] The temperature sensor judgment algorithm can perform judgment functions beyond simple calculations, and can reduce excessive false positives by suppressing the surge effect of outliers through logarithmic function-based normalization and offsetting repetitive operating patterns through periodic correction.

[0129] Compared to existing simple threshold comparison methods, the temperature sensor determination algorithm can reflect both the degree of agreement between sensors and environmental consistency, so it is possible to expect the effect of reducing the frequency of alarms under the same conditions and allowing for greater focus on genuine abnormal situations.

[0130] The temperature sensor determination algorithm simultaneously reflects the consistency of the output of sensor i with other sensors, consistency with the environment, correction for operating periodicity, and thresholds of external environmental conditions to form a single sensor normality index (F i It can be configured to extract (t).

[0131] The first element of the temperature sensor determination algorithm is the surface temperature measurement of sensor i (T si Representative value of other surface sensors of the battery, such as ) (T s Add the absolute value to the difference between '), and the difference to the reference temperature (δ Fi After standardizing the size by dividing by ), adding 1 to the result and taking the natural logarithm, the first logarithmic weight (k Fi1 The degree of agreement between sensors can be scored by multiplying by ).

[0132] The second element of the temperature sensor determination algorithm is the surface temperature measurement (T) of sensor i. si ) and ambient temperature (T a Add the absolute value to the difference of the ) values, and the difference is the reference temperature (δ Fi Standardize the size by ) and take the natural logarithm, then the second logarithmic weight (k Fi2 Consistency with the environment can be scored by multiplying by ).

[0133] The third element of the temperature sensor determination algorithm is the elapsed time (t Fi ) time scale (τ Fi Obtain the periodic oscillation value by applying the trigonometric function sin to the value divided by ), and to that value, the periodic correction coefficient (β Fi By multiplying by ) and subtracting from the total exponent, systematic fluctuations due to periodicity can be offset.

[0134] The fourth element of the temperature sensor judgment algorithm is the judgment threshold (θ) representing the external environmental conditions. Fi By subtracting ) from the total, a baseline can be established to recognize small deviations as normal.

[0135] Finally, the temperature sensor determination algorithm sums all four aforementioned elements to obtain the sensor normality index (F) of sensor i. i (t)) can be extracted, and if this index is 0 or less, it is considered normal, and if it exceeds 0, it is determined that the sensor requires inspection.

[0136] The temperature sensor determination algorithm uses the surface temperature measurement (T) of sensor i, which serves as the most direct basis for internal temperature estimation and sensor normality determination. si Using ) the sensor normality index (F i (t)) can be extracted.

[0137] Surface temperature measurement of sensor i (T si ) can use °C units and can be derived from values ​​obtained by sampling at intervals of 1 to 5 seconds through a thermistor, RTD, or thermocouple attached to the battery surface.

[0138] Surface temperature measurement of sensor i (T si The initial value of ) can be set to the value measured in real time after applying the calibration table of each sensor.

[0139] The temperature sensor determination algorithm uses the representative value (T) of other surface sensors that can serve as a baseline for determining the degree of agreement between sensors. s Using ') sensor normality index (F i (t)) can be extracted.

[0140] Representative value of other surface sensors (T s ') may refer to the average value collected at the same time from different sensors located at different positions on the same battery, and may use °C units.

[0141] The temperature sensor determination algorithm uses the ambient temperature (T) as a comparison criterion to verify whether the surface value is completely detached from the environment. a Using ) the sensor normality index (F i (t)) can be extracted.

[0142] Ambient temperature (T a ) can use °C units and can be derived from values ​​collected at equal intervals through an ambient temperature sensor installed at a designated point in the air layer around the battery or in the duct flow path.

[0143] Ambient temperature (T a ) can be set from the value collected after reflecting the calibration offset after sensor installation.

[0144] The temperature sensor determination algorithm manages the phase of the trigonometric function sin to be constant and reflects the elapsed time (t) to ensure stable operation of period correction. Fi Using ) the sensor normality index (F i (t)) can be extracted.

[0145] Elapsed time (t Fi ) can mean the elapsed time from the reference point (t0), can use the unit of 'seconds', and can be derived from the value obtained by reading the current time from the NTP-synchronized system time of the management server and subtracting the reference point from the current time.

[0146] The temperature sensor determination algorithm uses a reference temperature (δ) to safely reflect the unit-based temperature difference in a logarithmic function. Fi Using ) the sensor normality index (F i (t)) can be extracted.

[0147] Reference temperature (δ Fi ) can represent a reference temperature for nondimensionalizing log input, can use °C units, and can be set to 1°C as the initial value.

[0148] The temperature sensor determination algorithm uses either the first element reflecting a logarithmic function or the second element as the sensor normality index (F i To reflect whether it has more influence on (t)), the first log weight (k Fi1 ) and the second log weight (k Fi2You can utilize ).

[0149] 1st log weight (k Fi1 ) and the second log weight (k Fi2 ) can be used because the site conditions where the battery is located are not always the same, and in the case where many sensors are evenly distributed and airflow is even, and temperature sensing sensors read similar values ​​most of the time, the first log weight (k Fi1 ) is the second logarithmic weight (k Fi2 You can also enter a value greater than ).

[0150] 1st log weight (k Fi1 ) and the second log weight (k Fi2 ) represents the operator's judgment on which of the two signals to trust more and can be used as a dimensionless value.

[0151] 1st log weight (k Fi1 ) and the second log weight (k Fi2 ) can be entered as a value between 0 and 1 by the server administrator or user depending on the battery installation type and the number of temperature sensing sensors, and can also be set as a value derived using an AI model that has learned information about the site where the battery is installed.

[0152] For example, an AI model can refer to a regression model, etc., which is the surface temperature measurement value (T) of sensor i over time. si ), representative value of other surface sensors (T s '), ambient temperature (T a ) learns, and once learning is complete, the sensor normality index (F) extracted by reflecting the importance of the first element and the importance of the second element to a certain extent, respectively. i Determine whether the value of (t)) is a reliable value and the first log weight (k Fi1 ) and the second log weight (k Fi2 ) can be derived.

[0153] The temperature sensor determination algorithm uses a period correction coefficient (β) to offset systematic fluctuations caused by operating periodicity without excessively. Fi Using ) the sensor normality index (F i (t)) can be extracted.

[0154] Period correction factor (β Fi ) can be a dimensionless value representing a coefficient that determines the magnitude of the periodic correction term, and the algorithm can be utilized by initially setting it to a small value so that it is not significantly affected by elements containing trigonometric functions.

[0155] Period correction factor (β Fi The initial value of ) can be set to within half of the judgment threshold.

[0156] The temperature sensor determination algorithm uses a time scale (τ) for synchronization of period correction. Fi Using ) the sensor normality index (F i (t)) can be extracted.

[0157] Time scale (τ) Fi ) can be defined as a time scale, the unit can be set to seconds, and the value can be derived by roughly identifying the period of the repeating pattern in the driving log.

[0158] The temperature sensor judgment algorithm accepts small deviations as normal to reduce unnecessary alarms, using a judgment threshold (θ Fi Using ) the sensor normality index (F i (t)) can be extracted.

[0159] Judgment threshold (θ Fi The unit of ) is dimensionless and can be derived from a pre-entered value by considering the battery model and the location environment of the battery at the administrator terminal (300).

[0160] The temperature sensor determination algorithm can be characterized by being configured to solve three common problems that occur in the field.

[0161] First, since a single sensor alone can be susceptible to accidental noise or local radiation effects, the surface temperature measurement (T) of sensor i is used to compare multiple sensors. si ) and the representative value of other surface sensors (T s It can be characterized by using the difference of ') and applying a logarithmic function so that the penalty does not become excessively large even if a large difference appears.

[0162] Second, to distinguish between individual anomalies and collective frostbite even in situations of rapid external temperature change where all sensors may shake simultaneously, the surface temperature measurement of sensor i (T) is used to additionally reflect environmental standards. si ) and ambient temperature (T a It can be characterized by using the difference of ) and applying a logarithmic function so that even if a large difference appears, the penalty does not become excessively large.

[0163] Third, since a pattern of uniform value fluctuations at specific times may repeat due to the fan or pump driving cycle, elapsed time (t) is used to reflect this. Fi ) time scale (τ Fi The value input into the trigonometric function sin after normalization by dividing by the periodic correction coefficient (β Fi It can be characterized by being configured to subtract together with ).

[0164] [Table 2]

[0165]

[0166] As shown in [Table 2] above, normalization using a logarithmic function can achieve the objective of mitigating the influence of outliers, and weights can adjust the relative importance of the two signals according to the situation, and the judgment threshold (θ Fi ) can function as a simple and interpretable device that provides external environmental conditions.

[0167] As shown in [Table 2] and the description above, the temperature sensor determination algorithm may be characterized by being designed using functions such as logarithmic functions through examples of various tests, and may be characterized by being designed to reflect actual data distribution and experimental experience.

[0168] For example, the surface temperature measurement (T) of sensor i si ) is 12°C, and the representative value of other surface sensors (T s ') is 11.5°C, and the ambient temperature (T a ) is 10°C, and the reference temperature (δ Fi ) is 1°C, and the first logarithmic weight (k Fi1 ) is 0.7, the second logarithmic weight (k Fi2 ) is set to 0.3, and the period correction coefficient (β Fi ) is 0.3, time scale (τ Fi ) is 900 seconds, elapsed time(t Fi ) is 900 seconds, judgment threshold (θ Fi When ) is set to 0.8, the sensor normality index (F i (t)) can be extracted as approximately -0.439.

[0169] The temperature sensor determination algorithm determines the surface temperature measurement value (T) of sensor i at predetermined intervals. si ) and the representative value of other surface sensors (T s ') and ambient temperature (T a ) and elapsed time (t Fi It can collect ), and immediately after collection, it can clean up outliers by applying a moving average filter and standard deviation-based threshold removal, and the sensor normality index (F i (t)) can be extracted, and the extracted sensor normality index (F i (t)) can be implemented in a way that supports control judgment and recording by simultaneously transmitting it to the temperature correction module (170).

[0170] Sensor normality index (F iIf (t)) exceeds 0, information required for checking sensor i can be sent to the administrator terminal (300) as a notification.

[0171] The temperature sensor determination algorithm is based on the reference temperature (δ through the administrator terminal (300) Fi ), first log weight (k Fi1 ) and the second log weight (k Fi2 ), period correction factor (β Fi ), time scale (τ Fi ), judgment threshold (θ Fi It can be configured by receiving ) as input, and the sensor normality index (F) of sensor i i If (t)) exceeds 0, it may be indicated as requiring inspection, and the sensor normality index (F) of the same sensor i If (t)) exceeds 0 at several consecutive points in time, the accumulated confidence over time can be increased to raise the priority of the alarm.

[0172] In addition, when judgments occur simultaneously for multiple sensors, a "majority AND-OR" rule may be applied. For example, an OR-majority rule may be used to generate a pack-level alarm if two or more of the three sensors within the same battery exceed 0, while conversely, a strong alarm that directly intervenes in heating or cooling control may be configured as an AND rule that allows it only when a specific key position sensor exceeds 0 and the environmental deviation term is above a certain level.

[0173] The temperature sensor determination algorithm is calculated iteratively over time and can directly reflect the pattern of change.

[0174] The temperature sensor determination algorithm checks the sensor normality index (F) at a period of 1 to 5 seconds. i (t)) can be updated, and the sensor normality index (F) can be calculated by combining the mean and rate of change of the most recent N samples. i (t)) or the surface temperature measurement of sensor i (T si ) and the representative value of other surface sensors (Ts The difference in ') and the surface temperature measurement of sensor i (T si ) and ambient temperature (T a It can detect rapid changes where the difference of ) exceeds a threshold.

[0175] When a sudden change is detected, the temperature detection module (110) first sends a "caution" grade to the administrator terminal (300) and can be promoted to a "warning" grade when it persists for a certain period of time.

[0176] In addition, the temperature sensor determination algorithm uses a time scale (τ) on a daily basis or at the time of operation schedule change. Fi By re-evaluating ), the trigonometric function can be maintained in continuous synchronization with the actual period, and in seasons or driving modes where environmental noise increases, the period correction factor (β) Fi The stability of the judgment can be maintained by slightly lowering ).

[0177] The temperature sensor determination algorithm may be characterized by this time-based logic being designed to respond sensitively to long-term trend changes while suppressing short-term noise.

[0178] In addition, the temperature sensor determination algorithm extracts the sensor normality index (F) from the input value received from the administrator terminal. i It may also be configured to be predictably reflected in (t).

[0179] The manager has a judgment threshold (θ Fi You can lower alarm sensitivity by increasing ) and, conversely, the judgment threshold (θ Fi Sensitivity can be increased to detect even minute deviations by lowering ), and the first log weight (k Fi1 If you increase ), the weight of the first element increases, making it more sensitive to the sensor that stands out in the crowd.

[0180] In addition, the manager has a second log weight (k Fi2 By increasing ) and raising the weight of the second factor, abnormal alerts can be detected more actively during sudden changes in external conditions, and the period correction coefficient (βFi Reducing ) can be reflected to weaken the period correction, and the period correction coefficient (β Fi Increasing ) increases the cancellation of period correction, but to prevent excessive changes in the judgment direction, the judgment threshold (θ) Fi It may be recommended to maintain it within half of )

[0181] The manager is the time scale (τ) Fi By changing ), the period of the trigonometric function can be matched to the battery's operating pattern, and the reference temperature (δ Fi By changing ) to adjust the standardization scale of the logarithmic function, the sensitivity to small deviations can be adjusted.

[0182] Each variable and item of the temperature sensor judgment algorithm is specifically defined, measured, and configured to be summable without unit conflicts, thereby providing reliable results and achieving technical effects such as improved precision and effectiveness compared to existing methods, which can be clearly understood and implemented even by ordinary technicians.

[0183] The heating control module (130) can control the heating module (700) provided on one side of the battery to heat the battery at low temperatures and release heat.

[0184] The heating module (700) may refer to a module capable of generating heat to raise the internal temperature of the battery, including a PTC heater, a heat pad, and a Peltier.

[0185] The heating module (700) can continuously vary the amount of heat generated by a control command received from the heating control module (130).

[0186] The cooling module (900) can control a cooling module provided on one side of the battery to cool the battery at high temperatures and release cold air.

[0187] The above cooling module (900) may refer to a module capable of releasing cold air to lower the internal temperature of the battery, including a fan, pump, valve, liquid cooling, heat sink, and Peltier.

[0188] The cooling module (900) can continuously vary the amount of cooling by means of a control command received from the cooling control module (150).

[0189] The temperature correction module (170) can generate a control command to control the operation of the heating module (700) and the cooling module (900) so that a preset appropriate temperature range of the battery is maintained, and can transmit the generated control command to the heating module (700) or the cooling module (900).

[0190] The temperature correction module (170) can generate a control command to drive the heating module (700) and stop the operation of the cooling module (900) when the current internal temperature of a specific battery is lower than the preset appropriate temperature range of the battery, and transmit the command to the heating module (700).

[0191] The temperature correction module (170) can generate a control command to operate the cooling module (900) and stop the operation of the heating module (700) when the current internal temperature of a specific battery is higher than the preset appropriate temperature range of the battery, and transmit the command to the heating module (700).

[0192] If there is a battery that is outside the error range of the preset appropriate temperature range of the battery, the temperature correction module (170) can send the unique number (ID) of the battery, the current internal temperature of the battery, and the current time to the administrator terminal (300).

[0193] The error range of the pre-set optimal temperature range of the battery may refer to the range pre-entered by the administrator terminal.

[0194] Meanwhile, the temperature correction module (170) may utilize a normalization judgment correction algorithm to determine and correct the value of a control command generated to be transmitted to a heating control module or a cooling control module.

[0195] The normalization judgment correction algorithm reduces the influence of outliers by normalizing low temperature deviations and high temperature deviations with a logarithmic function, respectively, and reduces the transient response that occurs when the range is exceeded by applying braking based on the trigonometric function sin for the rate of change.

[0196] Additionally, the normalization judgment correction algorithm is a judgment control logic that operates within the temperature correction module (170) during the overall flow of the invention, and receives internal temperature and time information provided by the temperature sensing module (110) as input, calculates a control output, and performs the role of transmitting a continuous variable control command to the heating control module (130) and the cooling control module (150).

[0197] The normalization judgment correction algorithm can reduce the amplitude of excess vibration and the return time after reaching the target range compared to the simple threshold-based ON / OFF method.

[0198] For example, if the same load conditions are assumed in an external environment, when applying the normalization judgment correction algorithm, the excess vibration amplitude after returning to the target range can be reduced from 1.2°C to 0.4°C, and the return time can also be shortened.

[0199] The normalization judgment correction algorithm may include a first element that reflects compensation for deviations on the lower temperature side.

[0200] The first factor can increase in value only when the battery internal temperature is lower than the lower limit specified by the user.

[0201] The first element is based on the difference from the lower bound, but it can be characterized by being designed to increase gradually by applying a logarithmic function so that it responds delicately to small differences and does not jump excessively to large differences.

[0202] The normalization judgment correction algorithm may include a second element that reflects compensation for deviations on the high temperature side.

[0203] The second factor can increase in value only when the battery's internal temperature exceeds the upper limit.

[0204] The second element can also be characterized by being designed to apply a logarithmic function so that the response does not become excessively large even for large excess deviations.

[0205] The normalization judgment correction algorithm is configured to subtract a second element reflecting high temperature deviation compensation from a first element reflecting low temperature deviation compensation, and can be interpreted to mean that if the extracted control output value (u(t)) is positive, a control command is generated to control the heating module (700), and if it is negative, a control command is generated to control the cooling module (900).

[0206] In addition, the normalization judgment correction algorithm may include a third element that reflects braking on the rate of temperature change, and reflects how quickly the internal temperature rises or falls over time, so that if it is already rising rapidly, the heating output is reduced, and if it is already falling rapidly, the cooling output is reduced.

[0207] In this case, a stable input range can be created by normalizing the temperature change rate of the third element to the representative maximum change rate observed in the recent interval and clamping the ratio so that it does not exceed the range of -1 to +1.

[0208] Subsequently, the trigonometric function sin can be applied to the normalized temperature change rate value to correct for smooth reflection of the braking effect.

[0209] In summary, the temperature correction module (170) can extract a control output value (u(t)) using a normalization judgment correction algorithm that combines the three elements described above, and if this control output value (u(t)) is positive, it can be transmitted as a value to drive the heating module (700), and if it is negative, it can be transmitted as a value to drive the cooling module.

[0210] The normalization judgment correction algorithm uses the internal temperature value (T) as a key state variable directly compared with the target range. in The control output value (u(t)) can be extracted by utilizing ).

[0211] Internal temperature value (T in ) can represent an estimated value indicating the actual internal temperature of the battery, and the unit can be °C. It can be derived from an estimated value obtained by fusing time-series data collected from a surface temperature sensor attached to the outside of the battery and an ambient temperature sensor around the battery through a battery model (e.g., a simple first-order model based on thermal capacity or thermal conductivity).

[0212] Internal temperature value (T in The initial value of ) can be set to a value estimated by immediate estimation and then updated every period (e.g., 1 second, 5 seconds).

[0213] The normalization judgment correction algorithm uses a lower limit value (T) to clearly define the target conditions for temperature maintenance. min ) and upper limit value (T max It can be characterized by the application of ).

[0214] lower limit value (T min ) and upper limit value (T max ) may mean a boundary value of the target range that the administrator directly inputs as a number on the terminal (300), and the unit may be °C, and may be derived as a value input by the administrator of the management server (100).

[0215] lower limit value (T min ) and upper limit value (T maxThe initial value of ) is based on the recommended range presented in the product guide, but can be adjusted by the manager according to the field conditions.

[0216] Since the normalization judgment correction algorithm must reflect not only the current position (temperature) but also the direction and speed of movement to reduce transient response, the control output value (u(t)) can be extracted by utilizing the rate of change of temperature (T').

[0217] The rate of change in temperature (T') refers to the rate of increase or decrease of the internal temperature over time. Its unit can be °C / min (or °C / second), and it can be derived by calculating the difference between the current internal temperature and the internal temperature of the previous cycle by the time difference, and by applying a moving average to reduce noise.

[0218] The initial value of the temperature change rate (T') can be set to the value collected from the time the temperature sensing sensor acquires the second sample.

[0219] The normalization judgment correction algorithm normalizes the rate of change from -1 to +1 to stably limit the input of trigonometric functions using the representative maximum rate of change (v max It can be characterized by utilizing ).

[0220] Representative maximum rate of change (v) max ) can represent the representative maximum value obtained by taking the upper percentile (e.g., 95th) of the absolute values ​​of the rate of change observed over a recent period of time (e.g., 10 minutes, 30 minutes), and the unit can be used as °C / min.

[0221] Representative maximum rate of change (v) max ) can be derived by periodically accumulating the log of the rate of change to estimate the distribution and reading the specified percentile value.

[0222] It can be explained that the initial value uses a temporary default value (e.g., 1.0 °C / min) immediately after system startup and is automatically updated as data accumulates. The reason for this is to stably limit the sine function input by normalizing the rate of change from -1 to +1.

[0223] The normalization judgment correction algorithm ensures physical consistency by making the input of the logarithmic function independent of units, using a dimensionless criterion (τ). u It can be characterized by utilizing ).

[0224] dimensionless criterion (τ u ) can be used as a scale constant to convert the temperature difference into a unitless value, and the unit can be °C, or it can be set as a fixed constant or initially set to 1 °C according to the characteristics of each product family.

[0225] Since the required compensation magnitude for the normalization judgment correction algorithm varies depending on the thermal inertia of each device even with the same deviation, a temperature deviation weight (k) is used to easily match the output magnitude. ue It can be characterized by utilizing ).

[0226] Temperature deviation weight (k ue ) is a dimensionless constant used to control sensitivity to temperature deviation, and it can be adjusted and set according to the external environment in which the battery is located. The reason for using it can be explained as being to easily match the output size, as the required compensation size varies depending on the thermal inertia of each device even for the same deviation.

[0227] The normalization judgment correction algorithm applies rate of change weights (k) to achieve smooth control by appropriately applying braking according to the rising or falling speed. ud It can be characterized by utilizing ).

[0228] Change rate weight (k ud) is a dimensionless constant for controlling the braking strength of the rate of change, and can be adjusted by increasing the value if the overshoot is severe and decreasing the value if the response is sluggish.

[0229] As described above, the normalization judgment correction algorithm may be characterized by being designed to reduce unnecessary fine outputs by decreasing deviations within the range of each variable, and to increase stability by limiting the sine input to the range of -1 and +1.

[0230] [Table 3]

[0231]

[0232] As shown in [Table 2] above, the normalization judgment correction algorithm can utilize a logarithmic function that increases more slowly as the input increases, and can be characterized as being used to suppress excessive amplification in large deviations.

[0233] In addition, the normalization judgment correction algorithm can utilize trigonometric functions to create a smooth and continuous curve for inputs in the range of -1 to +1, and can be characterized by being used to smoothly dampen the rate of change.

[0234] As shown in [Table 2] and the description above, the temperature sensor determination algorithm may be characterized by being designed using functions such as logarithmic functions through examples of various tests, and may be characterized by being designed to reflect actual data distribution and experimental experience.

[0235] In addition, since the normalization judgment correction algorithm must be controlled around a manager-specified range for battery protection and performance maintenance, it may be characterized by being designed to apply a minimum value of 0 so that the first and second elements operate only when they go outside the range.

[0236] This can have the effect of creating a stable stop band by reducing unnecessary fine control within the range.

[0237] Next, the normalization judgment correction algorithm can be characterized by using a logarithmic function that reduces the growth rate as the input increases, as transient response and oscillation may increase if the output grows too quickly as the magnitude of the deviation increases.

[0238] In addition, since a response may be delayed if only the current temperature is considered, predictive control can be added by reflecting the rate of change to reduce the operation of the heating module (700) in advance if it is already rising rapidly, and reduce the operation of the cooling module (900) in advance if it is already falling rapidly.

[0239] At this point, the rate of change is normalized to the representative maximum rate of change and clamped to secure a stable input range of -1 and +1, after which the braking strength can be smoothly generated using trigonometric functions.

[0240] Finally, by introducing dimensionless constants for log inputs to eliminate unit dependency, consistent behavior can be achieved across different products or environments.

[0241] For example, lower limit value (T min ) is 15°C, upper limit value (T max ) is 30°C, representative maximum rate of change (v max ) is 1.0°C / min, dimensionless reference (τ u ) is 1°C, temperature deviation weight (k ue ) is 1.0, change rate weight (k ud If ) is set to 0.5, the internal temperature value (T in It can be assumed that ) is 12°C and is rising at a rate of 0.3°C per minute.

[0242] At this time, the difference from the lower limit can be calculated as 3°C, the compensation through the logarithmic function is approximately 1.386, the high temperature compensation is 0 since there is no exceeding of the upper limit, the rate of change ratio is 0.3, and the trigonometric function value is approximately 0.454, so the final control output value (u(t)) can be extracted as approximately 1.159, and this can be transmitted to generate a command to drive the heating module (700).

[0243] The normalization judgment correction algorithm collects the battery's surface temperature and ambient temperature to obtain the battery's internal temperature value (T in ) can be estimated, and the difference from the previous period value can be divided by the time interval to derive the rate of change of temperature (T'), and the rate of change of temperature (T') can be divided by the representative maximum rate of change to create a ratio, and then if it is outside the range of -1 to +1, it can be restricted by clamping.

[0244] Internal temperature value (T in ) lower value (T min ) and upper limit value (T max After calculating each deviation (negative deviations are converted to 0) by comparing with ), unit dependency is removed by dividing by a dimensionless constant, and then a logarithmic function is applied to the deviation and a sine function to the rate of change to obtain the values ​​of the three parts, and then sum them with a defined weight to extract the control output value (u(t)).

[0245] If the extracted control output value (u(t)) is positive, it can support generating a control command for driving the heating module (700), and if it is negative, it can support generating a control command for driving the cooling module (900).

[0246] The normalization judgment correction algorithm uses range conditions to determine that the low-temperature deviation compensation condition is satisfied if the internal temperature is lower than the lower limit, and the high-temperature deviation compensation condition is satisfied if it is higher than the upper limit.

[0247] The aforementioned range conditions can be mutually exclusive, and it can be explained that within the range, both deviation compensations can be zero.

[0248] The normalization judgment correction algorithm is a rate condition, and if the ratio of the absolute value of the rate of change to the representative maximum rate of change exceeds 1, it can be limited to 1 through clamping.

[0249] The aforementioned speed condition can be used as an essential condition for ensuring the stability of the sine function input.

[0250] The normalization judgment correction algorithm is a combinational method, where low-temperature deviation compensation and high-temperature deviation compensation can be combined as the difference between values ​​calculated through a logarithmic function, and the rate of change damping term can be added or subtracted with a separate weight to become the total control output.

[0251] The normalization judgment correction algorithm can have a deadband as an auxiliary criterion that considers the absolute value of the final output to be zero when it is very small (e.g., less than 0.05), thereby reducing driving vibration.

[0252] The normalization judgment correction algorithm can adapt to environmental changes by allowing the lower and upper limits to be adjusted frequently via administrator input due to the fluidity of the reference value, and by automatically updating the representative maximum rate of change based on data accumulation.

[0253] The normalization judgment correction algorithm can update the internal temperature and rate of change through periodic sampling (e.g., at 1-second intervals, 5-second intervals).

[0254] If it is out of range at a specific point in time, the first or second element reflecting deviation compensation can be immediately activated, and at the same time, the third element based on the rate of change of trigonometric functions can work together to suppress excessive heating or heat loss.

[0255] As time passes and the internal temperature returns to the target range, the deviation compensation term can naturally decrease to zero, and as the rate of change decreases, the damping term also decreases and can converge to a steady state.

[0256] In addition, if the distribution of the rate of change in the recent section changes, the representative maximum rate of change is newly updated, and the subsequent braking action can be automatically rescaled to match the field conditions.

[0257] Each variable and item of the normalization judgment correction algorithm is specifically defined, measured, and configured to be summable without unit conflicts, thereby providing reliable results and achieving technical effects such as improved precision and effectiveness compared to existing methods, which can be clearly understood and implemented even by ordinary technicians.

[0258] A battery system for heavy equipment that supports wireless charging according to the present invention is configured to include a power transmitter (100) installed on the ground or a specific structure to supply power wirelessly, and a power receiver (200) mounted on heavy equipment to receive power from the power transmitter (100) and charge a battery. The power transmitter (100) receives power from an external power source to form a magnetic field and performs the role of transmitting power to the power receiver (200), and for this purpose includes a power transmission unit (110) and a transmission coil unit (120). The power transmission unit (110) includes an inverter circuit that receives commercial AC power and converts it into AC power having a resonant frequency. The power transmission unit (110) variably controls the magnitude or frequency of the transmitted power based on battery status information received from the communication module (260) of the power receiver (200). The transmitting coil unit (120) is electrically connected to the power transmitting unit (110) and is composed of a primary coil that generates a time-varying magnetic field by alternating current power applied from the power transmitting unit (110). The power receiving unit (200) is installed at the bottom of the heavy equipment or near the battery pack and receives power wirelessly from the power transmitting unit (100) to charge the battery. To this end, the power receiving unit (200) includes a receiving coil unit (210), a power conversion unit (220), a battery management unit (230), a central control unit (240), a charging control module (250), a communication module (260), and a position alignment module (270). The receiving coil unit (210) is composed of a secondary coil that generates an induced electromotive force by electromagnetically coupling with the time-varying magnetic field generated from the transmitting coil unit (120). The power conversion unit (220) converts the alternating current power received through the receiving coil unit (210) into direct current power suitable for charging the battery of heavy equipment, and for this purpose includes a rectifier circuit and a DC-DC converter circuit.The battery management unit (230) is provided inside the battery pack and measures and monitors the voltage, current, temperature, and charge status of the battery cells in real time. If the measured battery information deviates from a preset reference value, the battery management unit (230) transmits a warning signal to the central control unit (240) to perform a protection operation. The central control unit (240) is composed of a microprocessor device that controls the overall operation of the power receiving unit (200). The central control unit (240) receives battery information from the battery management unit (230) and executes the charging control module (250), the communication module (260), and the position alignment module (270) to comprehensively control the wireless charging process. The charging control module (250) is software logic executed by the central control unit (240) and controls the power conversion unit (220) to sequentially perform a constant current charging step and a constant voltage charging step according to the battery status transmitted from the battery management unit (230). The communication module (260) transmits and receives data with the power transmitter (100) using a short-range wireless communication method. The communication module (260) transmits a charging start request, real-time battery status information, a charging completion signal, etc., to the power transmitter (100) to ensure that the charging procedure is performed stably through mutual information exchange. The position alignment module (270) is composed of software logic that determines the alignment accuracy between the transmitting coil unit (120) and the receiving coil unit (210). The position alignment module (270) determines whether the horizontal separation distance between the center axes of the two coils is located within a pre-set allowable error range based on a magnetic field sensor or the strength of the received power. Charging is initiated only when the allowable error range is satisfied, thereby allowing charging to proceed when the power transmission efficiency is secured above a certain level.The central control unit (240) performs a 'battery risk state judgment process' that comprehensively evaluates multiple heterogeneous data rather than a single physical quantity to diagnose the safety of the battery in real time. The process is configured to use three key judgment criteria. First, battery temperature, which refers to the temperature of the internal cells of the battery, is judged to evaluate the current thermal stability of the battery and detect imminent danger. Second, gas leakage concentration, which refers to the concentration of electrolyte vapor or specific gas generated when the battery cells are internally damaged or overcharged or overdischarged, is judged to predict potential danger by detecting early whether physical damage or chemical abnormal reactions have occurred. Third, internal resistance, which is a resistance value for charge movement within the battery and tends to increase when the battery ages or internal micro-short circuits occur, is judged to diagnose long-term state changes and potential internal defects of the battery. Of course, the battery management unit (230) monitors the voltage and current of the battery cells in real time and performs basic protection logic to block electrical abnormal states such as overcharging, overdischarging, and overcurrent. However, the 'battery risk state determination process' of the present invention goes beyond such basic electrical protection and is characterized by its technical features in predicting and detecting warning signs of fatal failures that are irreversible once they occur, such as thermal runaway. Accordingly, the temperature, gas, and internal resistance, which directly indicate changes in the physicochemical state of the battery, are adopted as key determination criteria. The system does not adopt a method of simply summing or averaging the three determination criteria. This is based on the technical reasoning that the type and urgency of the risk indicated by each criterion differ. For example, a gradual increase in internal resistance is judged as a 'long-term' signal requiring maintenance, whereas a rapid simultaneous rise in temperature and gas concentration is judged as a 'short-term' and fatal risk signal requiring immediate action.Accordingly, the system comprises a logical processing procedure based on hierarchical priority rules. The procedure first performs a primary evaluation by independently evaluating each judgment criterion to assign an individual risk grade. Subsequently, it performs a secondary evaluation to determine the final risk state by combining the grades and assigning a higher weight or priority to the grade of the judgment criterion with a high urgency of risk. The central control unit (240) acquires data for each judgment criterion in the following manner. The battery temperature is measured through multiple NTC thermistors installed inside the battery pack, and the unit is Celsius. Since the battery temperature shows rapid changes within a few seconds when a thermal runaway phenomenon occurs, the shortest measurement cycle of 1 second is set to ensure that the dangerous situation is not missed. For example, it is set so that it can be measured at a value of 62 Celsius. The gas leakage concentration is measured through a MOS type gas sensor for detecting electrolyte vapor installed within the sealed space of the battery pack, and the unit is ppm. Since it takes a certain amount of time for the gas generated by internal damage to reach the sensor and for the concentration to change, a measurement cycle of 5 seconds is set to compromise between securing a stable measurement value of the sensor and the response speed. For example, it is set so that the value can be measured at 150 ppm. The internal resistance is calculated by the battery management unit (230) applying a current pulse during charging or discharging and measuring the voltage drop that occurs at that time, and the unit is milliohms. Since the accuracy of the internal resistance is highest when measured when the battery is in an electrically stable state, it is not measured at a fixed cycle, but is calculated non-periodically under specific conditions where the central control unit (240) determines the charge state and stability of the battery and decides that it is most suitable for measurement. For example, it is set so that the value can be calculated at 35 mΩ. The final determination result of the 'battery danger state determination process' is classified into four stages: 'normal', 'caution', 'warning', and 'urgent', and a control action linked to each state is triggered.If determined to be in a 'Normal State,' all charging and operation procedures are performed normally. If determined to be in a 'Caution State,' the maximum charging current is reduced by a predetermined ratio via the 'Dynamic Charging Current Control Process,' and the relevant data is recorded in the system log. This reduction ratio is set by considering a safety margin derived from prior battery thermal characteristic simulation results or multiple experimental data, within a range that lowers the thermal load applied to the battery to a meaningful level without causing unnecessary charging interruptions. If determined to be in a 'Warning State,' charging is immediately stopped, and an audiovisual warning is transmitted to the heavy equipment operator. If determined to be in an 'Emergency State,' the 'Emergency Detachment Drive Control Process' is triggered simultaneously with the charging stoppage and warning issuance to immediately execute the physical disconnection of the battery pack. This is explained using an example scenario where minor internal damage occurs to some battery cells due to external impact while the heavy equipment is wirelessly charging. First, the central control unit (240) collects data from each sensor regarding a battery temperature of 50°C, a gas leak concentration of 120 ppm, and an internal resistance of 25 mΩ after 1 minute has elapsed since the damage. Second, the collected data is compared with a preset threshold value to determine the risk level of each criterion. The battery temperature of 50°C is determined as a 'Caution' level, the gas leak concentration of 120 ppm as a 'Caution' level, and the internal resistance of 25 mΩ as a 'Normal' level. Third, the combination of levels is evaluated according to a preset state determination rule table. Since it corresponds to the second rule, "If there are two or more 'Caution' level factors, it is determined as a 'Warning' state," the final 'battery risk state' is determined as 'Warning'. Fourth, a control operation corresponding to the 'Warning' state is performed to immediately stop wireless charging and output a warning sound along with a message "Battery overheating check required" on the driver's seat display. The central control unit of the present invention determines the final risk state according to the following priority rules.

[0259] State determination rule table Rule No. condition Final risk state Rule 1 If either the temperature or gas rating is 'Warning' emergency Rule 2 If it does not fall under Rule 1 and there are two or more 'Caution' grade factors warning Rule 3 If it does not fall under Rules 1 and 2, and there is only one 'Caution' grade factor caution Rule 4 If none of the above rules apply normal

[0260] The judgment logic of the present invention includes a sensor value validity verification step to prepare for exceptional situations such as sensor failure. When collecting data, the central control unit (240) first checks whether each sensor value is within a physically possible range. If a specific sensor value falls outside the range or communication is interrupted, the sensor is judged to be 'fail'. If one sensor is judged to be 'fail', the system determines the dangerous state using only the remaining valid sensor values ​​for safety, but responds conservatively by upgrading the final result by one level, or immediately sends a 'sensor check' warning to the driver. Compared to prior art, in a scenario involving a gradual temperature rise and gas leak due to an internal short circuit, prior art using temperature as a single criterion does not take any action until the temperature reaches 60°C. On the other hand, the present invention determines the state as 'warning' at the initial point when the temperature exceeds 45°C and the gas leak concentration exceeds 100 ppm, preemptively stops charging, and sends a warning. In a scenario involving an increase in internal resistance due to aging, prior art does not detect danger because the temperature does not rise. On the other hand, the present invention determines the state as 'Caution' when the internal resistance exceeds 30 mΩ, slightly reduces the charging current, and indicates the need for maintenance. In a scenario involving a rapid increase in temperature and gas levels due to external shock, the prior art merely cuts off charging when the temperature reaches 60°C, whereas the present invention determines the state as 'Emergency' immediately when the temperature exceeds 60°C and the gas leakage concentration exceeds 200 ppm simultaneously, thereby executing the physical disconnection of the battery. The central control unit determines the final danger state according to a state determination rule table. The first rule of the table determines the final danger state as 'Emergency' if either the temperature or the gas level is 'Warning'. The second rule determines the state as 'Warning' if it does not correspond to the first rule and there are two or more 'Caution' level factors.Rule 3 determines the state as 'Caution' if it does not fall under Rules 1 and 2 and there is only one 'Caution' grade factor. Rule 4 determines the state as 'Normal' if it does not fall under any of the above rules. Through simulation examples, when the temperature is 58°C ('Caution'), gas is 210 ppm ('Warning'), and resistance is 25 mΩ ('Normal'), the final danger state is determined as 'Urgent' according to Rule 1, whereas prior art fails to detect danger. The judgment rules presented in the present invention are composed of objective and inevitable elements based on a technical understanding of the physicochemical mechanisms of battery failure. Since substantial safety can only be ensured by early detection of complex failure scenarios that cannot be detected by a single factor alone, and by varying the level of response according to the severity of the risk, a combination of multiple judgment criteria and hierarchical rules is adopted as an essential configuration. By adopting multiple judgment criteria and hierarchical rules as described above, the present invention achieves the technical effects of ensuring predictive safety, improving the precision of response, and preventing secondary damage. Specifically, it detects warning signs to predict failures and responds preemptively, subdivides risk conditions and performs differential measures appropriate to each stage, and fundamentally prevents secondary damage that could spread to a fire throughout the entire vehicle by physically disconnecting the battery in an 'emergency' state. The threshold values ​​used in this specification are exemplary, and their criteria are set as follows. The basis for setting each threshold value is primarily based on the type of battery cell used and the safe operating range specified in the technical specifications provided by the manufacturer. This implies that the initial setting of the threshold values ​​is based on objective engineering data. The above threshold values ​​are not fixed values; rather, they are designed to be periodically updated via firmware updates to threshold values ​​optimized for a specific model or operating environment by statistically analyzing multiple heavy equipment battery data collected in actual operating environments.Specifically, the above update analyzes the distribution of specific sensor data accumulated in the system and executes the update procedure based on new statistical values, subject to administrator approval, when a specific percentile value of the collected data consistently shows a difference exceeding a predetermined error range from the previously set 'Caution' grade threshold. The risk grade determination criteria are set as follows: for temperature, Normal (45°C); Caution (over 45°C to 60°C); for gas, Normal (100 ppm); Caution (over 100 ppm to 200 ppm); and Warning (over 200 ppm); and for resistance, Normal (~30 mΩ) and Caution (over 30 mΩ).

[0261] Definition of threshold for risk level determination (Example) Judgment criteria unit Normal grade Caution level Warning level Basis for setting Battery temperature ℃ 45 or less Over 45 ~ 60 or less Over 60 NCM Battery Manufacturer Safety Specifications Gas leak concentration ppm 100 or less Over 100 ~ 200 or less Over 200 Electrolyte (LiPF6) Vapor Hazard Data Internal resistance mΩ 30 or less Over 30 (Not applicable) Battery State of Health (SOH) 80% baseline value

[0262] Risk assessment simulation results by scenario (compared to prior art) scenario Temperature (°C) gas (ppm) Resistance (mΩ) Judgment result of the present invention Result of prior art assessment (based on a temperature of 60℃) 1. Normal charging 42 30 18 normal normal 2. Internal micro-short circuit 55 150 25 Warning (2 Caution) Normal (risk detection failed) 3. Cell aging 40 25 35 Caution (1 Caution) Normal (Status diagnosis failed) 4. External impact 62 210 28 Urgent (2 warnings) Danger detection (charge cutoff)

[0263] Table 7 above illustrates the criteria for determining the risk level according to an embodiment of the present invention. Each threshold value is set based on objective data specified in the battery type and the manufacturer's technical specifications. Table 8 above shows the simulation results comparing the judgment logic of the present invention with the prior art under various virtual scenarios. As seen in Scenario 2, the prior art fails to detect danger because the temperature does not reach 60°C, whereas the present invention demonstrates that it can respond preemptively by determining the temperature and gas as 'Warning' because both correspond to the 'Caution' level. When the battery replacement unit (300) receives an 'emergency' status signal from the 'battery risk status judgment process,' it performs an 'emergency detachment drive control process' to quickly and safely physically separate the battery pack from the heavy equipment. The process does not simply operate at maximum output, but utilizes the following key judgment criteria for precise control. First, the target detachment time, which is the maximum allowable time for battery disconnection to be completed after an 'emergency' state occurs, is determined to secure a safety golden time to prevent secondary damage such as thermal runaway propagation. Second, the battery mass, which is the total mass of the battery pack to be detached, is determined to determine the basic driving force required to overcome inertia and obtain the necessary acceleration within the target time. Third, the required detachment distance, which is the minimum linear distance the battery must travel to ensure a thermally safe distance and to completely separate it from the electrical connection of the heavy equipment, is determined to optimize energy usage by limiting the displacement in the driving force calculation and to prevent additional damage caused by unnecessary movement. Fourth, the actuator specifications, such as the total number of actuators used to push out the battery and the piston cross-sectional area of ​​each actuator, are determined to determine the precise pressure or voltage to be applied to individual actuators by distributing the calculated total required driving force to each actuator.Fifth, frictional resistance—the static and dynamic frictional forces generated when the battery detaches from the fixed device and moves along the path—is determined to compensate for all physical resistances resisting actual operation, in addition to the battery's inertia, thereby preventing detachment failure. The system avoids adopting a simple mechanical method of pushing the battery with a fixed force. The technical rationale for this is that a fixed force cannot guarantee the target detachment time in real-world situations where battery masses differ or frictional resistance varies. The system constructs a logical processing procedure based on fundamental dynamic principles, such as Newton's laws of motion and the equation for uniformly accelerated motion. The process first back-calculates the acceleration required to move the battery mass by the required detachment distance within the target detachment time. Using the calculated acceleration and battery mass, the required net force is determined. The force required to overcome frictional resistance is added to this to determine the final total required driving force. Finally, this total required driving force is converted into pressure or voltage signals that each actuator must generate, utilizing the actuator specifications. In particular, the above total required driving force is determined by reflecting a physical relationship in which it does not increase linearly as the target detachment time decreases, but increases exponentially in inverse proportion to the square of the time.

[0264] Definition of parameters for emergency detachment control (Example) parameters sign Example values Basis for setting Target removal time (doesn't exist) 1.5 seconds Experimental data on thermal runaway cell propagation time Battery mass (doesn't exist) 300 kg Battery Pack Design Specifications by Heavy Equipment Model Distance required for removal (doesn't exist) 0.5 m Electrical / Thermal Safety Clearance Design Standards Maximum safety limit (doesn't exist) (e.g., 50 bar) Heavy equipment frame structure analysis results

[0265] Comparison of detachment success by condition (compared to prior art) scenario condition Prior art (fixed force) The present invention (dynamic control) 1. Standard conditions Mass 300kg, friction 500N Success (takes 1.4 seconds) Success (Takes 1.5 seconds, uses optimal force) 2. Weight increase Mass 400kg, friction 500N Failure (lack of strength) Success (Takes 1.5 seconds, stronger force calculated) 3. Increased friction Mass 300kg, friction 800N Failure (lack of strength) Success (Takes 1.5 seconds, power amplified by feedback)

[0266] As can be seen in [Table 10] above, prior art using a fixed force fails to achieve safety objectives because it fails to detach under conditions that deviate from design standards (increased weight, increased friction). On the other hand, the present invention demonstrates that detachment can be completed within the target time under any conditions by dynamically recalculating the required driving force by reflecting changed conditions as parameters or by amplifying the force through feedback. The control unit of the battery replacement unit (300) acquires and stores data for each judgment criterion in the following manner. The target detachment time is pre-set as a critical time to ensure safety based on experimental data regarding the battery thermal runaway propagation speed and is stored in the non-volatile memory of the control unit. For example, it is stored as 1.5 seconds. The battery mass, required detachment distance, and actuator specifications are fixed values ​​corresponding to the mechanical design specifications of the heavy equipment model and the mounted battery pack, and are pre-stored in the non-volatile memory of the control unit. For example, the battery mass is configured to be 300 kg, the required detachment distance is 0.5 m, there are two actuators, and the cross-sectional area of ​​each cylinder is 0.002 m². The frictional resistance is treated as a variable value rather than a fixed value. Although the initial frictional resistance value is set through design simulation, during system maintenance mode or periodic self-diagnosis, the frictional resistance value is measured and updated by having the actuator move the battery slightly at a very slow speed and measuring the minimum driving force required at that time. For example, it is configured to be measured as a pressure corresponding to 500 N. The final determination result of the above 'emergency detachment drive control process' is an 'actuator drive signal' corresponding to a specific pressure or voltage. The signal may take the form of a pulse width modulated signal or a DC voltage signal of a specific magnitude. The generated drive signal is transmitted to a hydraulic pump controller or a pneumatic valve controller so that the actuator accurately generates the determined driving force.Through this, a highly reliable physical measure is executed to always complete the battery disconnection within the target disconnection time, even under various conditions. This is explained by taking an example scenario where an 'emergency' status signal is generated and a 300 kg battery must be disconnected by 0.5 m within 1.5 seconds. First, an 'emergency' status signal is received from the central control unit (240). Second, parameters required for emergency disconnection are loaded from the control unit memory. The parameters are set as a target disconnection time of 1.5 seconds, a battery mass of 300 kg, a required disconnection distance of 0.5 m, two actuators, a cylinder cross-sectional area of ​​0.002 m², and a measured frictional resistance of 500 N. Third, the total required driving force is determined based on the parameters and the built-in dynamics-based logic. The control unit first determines the acceleration required to move 0.5 m in 1.5 seconds. Subsequently, the force required to overcome inertia is determined by applying the mass of 300 kg to this acceleration. Here, a frictional resistance of 500N is added to determine the total driving force required. Fourth, the determined total driving force is divided among two actuators, and the pressure required for each actuator to generate the corresponding force is calculated. A control signal corresponding to the calculated pressure is generated and automatically transmitted to a hydraulic or pneumatic system. The control logic ensures reliability by including closed-loop control through feedback rather than simple open-loop control. The battery replacement unit (300) is additionally equipped with a linear encoder or limit switch to detect the actual position of the battery. After transmitting the driving signal, the control unit monitors in real time whether the battery is actually moving through a position detection sensor. If no movement of the battery is detected even after a certain percentage of the target removal time has elapsed, it is determined that the frictional resistance has increased rapidly due to the insertion of an unexpected obstacle, etc. In this case, the system immediately switches to a primary safety mode and transmits a command to retry by amplifying the driving pressure to the maximum safety limit that the system can withstand.Compared to prior art, the prior art using a fixed-force spring method succeeds in detachment under standard condition scenarios, whereas the present invention succeeds in detachment by calculating precise force and avoids causing unnecessary shock to the equipment. In scenarios where a heavier battery is replaced, the prior art fails to detach or misses the golden time, whereas the present invention receives the changed battery mass value as a parameter and automatically calculates a stronger driving force to successfully detach within the target time. In scenarios where frictional resistance increases due to the insertion of foreign matter, the prior art fails to detach, whereas the present invention detects the absence of movement through position sensor feedback and immediately amplifies the pressure to the maximum safety limit to retry the detachment, thereby maximizing the success rate. The dynamic drive control method of the present invention is an essential configuration for solving the technical challenge of "ensuring successful physical separation within a set safety time under any conditions." To respond to physical variables that may vary depending on the type of battery, aging condition, and external environmental factors, it is essential to adopt dynamic control logic that precisely calculates and applies a force appropriate to the situation, rather than a fixed force. By adopting judgment criteria and control rules based on dynamic principles as described above, the present invention achieves technical effects of maximizing detachment reliability, ensuring equipment stability, and enhancing system scalability and adaptability. Specifically, it maximizes the success rate of detachment within a target time even under various battery specifications and situations involving unexpected increases in frictional resistance, prevents secondary damage caused by excessive impact, and enables immediate response by simply updating parameter values ​​even if the battery pack is replaced with one of a different mass in the future. The criteria for the core threshold values ​​used in this specification are defined as follows. The target detachment time is defined as an engineering safety time based on multiple experimental data regarding the average time required for thermal runaway in a lithium-ion battery to begin and propagate to adjacent cells, with a sufficient safety margin to ensure that detachment is completed before propagation occurs.The above maximum safety limit is the maximum value of pressure used during exception handling, and is determined by the pressure inversely calculated from the maximum load that does not cause plastic deformation or breakage through structural analysis of the battery replacement unit and the heavy equipment frame. The above central control unit (240) performs a 'charging abnormal state detection process' beyond simply comparing input and output power to diagnose the stability and efficiency of the wireless charging process in real time. The above process utilizes the following key judgment criteria to detect unpredictable failure factors. First, the transmission power, which is the real-time amount of power supplied from the power transmission unit (100) to the transmission coil unit (120), is determined to be used as a reference value defining the total input energy of the charging system. Second, the actual charging power, which is the real-time amount of DC power finally applied to the battery after passing through the power conversion unit (220) of the power reception unit (200), represents the final output energy of the system and is determined to be used as a key comparison target for efficiency and abnormal state judgment. Third, the coil alignment error, which is the horizontal and vertical separation distance between the central axes of the transmitting and receiving coils, quantifies one of the most common and predictable loss factors occurring in wireless power transmission, thereby determining the normal range of efficiency degradation. Fourth, the coil and battery temperature, which is a temperature value reflecting the increase in conductor resistance of the transmitting and receiving coils and the increase in internal resistance of the battery due to heat generation during charging, quantifies the power loss inevitably caused by the temperature rise, thereby determining the normal range of efficiency degradation. The above system does not adopt a method that simply calculates the final efficiency and compares it with a fixed threshold value. The technical basis for this is that it is a normal phenomenon for the efficiency of wireless charging to change moment by moment depending on the coil alignment state or ambient temperature. The above system configures a logical processing procedure that separates and determines 'predictable normal loss' and 'unpredictable abnormal loss'.The first step of the above procedure first calculates the 'expected power loss' that should normally occur under current conditions based on the current coil alignment error and temperature. Subsequently, the 'expected charging power' that should be achieved under current conditions is dynamically calculated by subtracting this expected power loss from the transmission power. The second step of the above procedure calculates the 'error rate' of the difference by comparing the calculated 'expected charging power' with the actually measured 'actual charging power'. The third step of the above procedure determines whether this 'error rate' exceeds a pre-set fixed threshold. Since the error rate refers to 'unidentified loss' caused by unpredictable causes such as the ingress of metal foreign matter between coils, damage to some components of the transmission / reception circuit, or rapid aging of the battery, it allows for the precise detection of actual abnormal conditions. The central control unit (240) acquires data for each judgment criterion in the following manner. The transmission power and actual charging power are measured in real-time at a 1-second interval through the power measurement circuits of the power transmission unit (100) and the power reception unit (200), respectively. For example, the transmission power is measured as 3500W and the actual charging power as 2950W. The coil alignment error is calculated by the position alignment module (270) analyzing the magnetic field signal strength pattern at the start of charging to determine the horizontal error between the center axes in mm units. For example, the horizontal error is calculated as 15mm. The coil and battery temperatures are measured at 1-second intervals through temperature sensors attached to each component. For example, the coil temperature is measured as 55℃ and the battery temperature as 42℃. The final determination result of the 'charging abnormal state detection process' is classified into a binary state of 'normal charging' or 'charging abnormal'. If determined to be in a 'normal charging' state, the charging procedure continues. If determined to be in a 'charging abnormal' state, wireless charging is immediately stopped, and a specific warning message such as "Charging system inspection required: Foreign object or system error detected" is automatically sent to the heavy equipment operator.The following is explained as an example of a scenario in which charging begins without noticing that a small metal tool has fallen onto the transmitting coil while heavy equipment is parked. First, the central control unit (240) collects data of a transmitting power of 3500W, an actual charging power of 2400W, a coil alignment error of 10mm, and a coil temperature of 50℃. Second, the control unit refers to a pre-stored lookup table to confirm that the normal loss rate corresponding to the current conditions is 15%, and based on this, calculates the 'expected charging power' as 2975W. Third, it compares the 'expected charging power' (2975W) with the 'actual charging power' (2400W) to calculate that the error rate is approximately 19.3%. Fourth, since the calculated error rate exceeds a pre-set allowable error threshold of 10%, the system finally determines this as a 'charging abnormality' state, immediately stops charging, and sends a warning to the operator. The detection logic includes a temporal filtering technique to prevent false positives. Even if the above 'error rate' exceeds a threshold, it is not immediately determined to be a 'charging abnormality.' Instead, it is finally confirmed as a 'charging abnormality' only when the state in which the error rate continuously exceeds the threshold is maintained for a preset period of time. Compared to prior art, in a scenario where the actual efficiency is 83% due to slightly poor coil alignment, the prior art, which is based on a fixed efficiency of 85%, misinterprets this as a 'failure' and stops charging. On the other hand, the present invention calculates that the 'expected efficiency' under the current condition is 84%, and since the actual efficiency of 83% is within the error range, it determines it as 'normal' and maintains charging. In a scenario where the efficiency is 90% due to circuit damage despite perfect coil alignment, the prior art misinterprets this as 'normal' and continues charging. On the other hand, the present invention calculates that the 'expected efficiency' under the current condition is 95%, and since the actual efficiency of 90% is outside the error range, it accurately detects it as a 'charging abnormality' and stops charging.The 'expected power'-based comparison method of the present invention is an inevitable configuration based on the engineering principle that precise fault diagnosis is possible only when losses occurring in a wireless charging system are clearly distinguished into 'normal losses that are predictable by design' and 'unpredictable abnormal losses caused by failure.' By adopting a judgment rule based on dynamic expected power as described above, the present invention achieves technical effects such as improved precision in fault diagnosis, enhanced safety, and increased convenience in maintenance. Specifically, it significantly reduces false positives and missed detections, accurately detects critical abnormal conditions that can cause overheating and fire at an early stage, and induces effective follow-up measures.

[0267] Simulation results for determining charging abnormality by condition (compared to prior art) scenario condition Actual efficiency Prior art assessment (based on 85%) The present invention 1. Normal (Perfectly aligned) Alignment error 5mm, temperature 40℃ 94% Normal (94% > 85%) 95% 2. Normal (slightly defective) Alignment error 30mm, temperature 50℃ 83% Failure (83% < 85%) 84% 3. Anomaly (Circuit Damage) Alignment error 5mm, temperature 40℃ 89% Normal (89% > 85%) 95% 4. Abnormalities (Ingress of foreign substances) Alignment error 10mm, temperature 55℃ 75% Failure (75% < 85%) 92%

[0268] Table 11 above compares the judgment results of the prior art using the dynamic judgment logic of the present invention and a fixed threshold under various scenarios. In Scenario 2, when the coil alignment is slightly poor and the efficiency drops to a normal 83%, the prior art mistakes this for a failure and causes unnecessary charging interruption. On the other hand, the present invention dynamically calculates that the 'expected efficiency' under the current conditions is 84%, and since the actual efficiency of 83% is within the margin of error, it determines it as 'normal' and demonstrates that stable charging can be maintained.

[0269] In addition, in Scenario 3, even though the alignment state is perfect, if the efficiency drops slightly to 89% due to internal circuit damage, the prior art mistakes this for normal and leaves potential risks unaddressed. On the other hand, the present invention demonstrates that it can accurately identify a 'charging abnormality' by detecting a significant difference (assuming an error rate of 6.3% and an allowable error exceeding 5%) that falls short of the expected efficiency of 95%. The criteria for the key threshold values ​​used in this specification are defined as follows. The lookup table is constructed based on empirical data obtained by extracting electromagnetic field simulations during the product design phase or by performing repetitive charging experiments with actual mock-up products. The allowable error threshold is an engineering tolerance set to satisfy a confidence level in which deviations exceeding the range can be considered as actual abnormal conditions, by statistically analyzing the maximum error range that may occur in a normal system while considering the accuracy of the prediction model and the measurement error of the actual system. The charging control module (250) performs a 'dynamic charging current control process' to simultaneously ensure the safety and long-term lifespan of the battery, going beyond simply charging the battery quickly. The above process utilizes the following key judgment criteria to determine the optimal charging current in accordance with the battery status changing in real time. First, the battery risk state, which is the overall safety status of the battery determined in real time through the above 'Process 1', sets battery safety as the top priority of all control logic and determines to immediately reduce or stop the charging intensity when signs of danger occur. Second, the current charge state, which is a percentage value indicating how much the battery's current charging capacity is relative to the total capacity, determines to implement a standard constant current-constant voltage charging profile that regulates the current according to the battery's charging stage and prevent overload in the full charge region.Third, the maximum allowable battery current—which is the maximum current value guaranteed by the battery manufacturer to allow safe charging without cell damage—sets an upper limit for all dynamic controls, ensuring that no current exceeding the battery's electrical specifications is applied under any circumstances. The system avoids adopting a simple constant current-constant voltage charging method that charges with a fixed constant current and switches to a constant voltage once a specific voltage is reached. The technical basis for this is that existing methods fail to reflect real-time changes in the battery's safety status during charging. The system comprises a logical processing procedure called "safety-first hierarchical control." The upper control layer of this procedure prioritizes the "battery dangerous state" as the primary judgment criterion. If the dangerous state is not "normal," this upper control layer directly and forcibly lowers or cuts off the charging current. The lower control layer of the above procedure is activated only when in a 'normal' state without separate intervention from the upper control layer, and gradually adjusts the charging current according to a traditional constant current-constant voltage charging profile based on the 'current charging state'. The charging control module (250) is configured to acquire data for each judgment criterion in the following manner. The battery risk state is directly received as one of the state values ​​'normal', 'caution', 'warning', or 'urgent' at a 1-second interval from the 'battery risk state judgment process' of the central control unit (240). The current charging state is calculated in real-time by the battery management unit (230) by accumulating the charging and discharging currents over time, and receives a percentage value updated at a 1-second interval. For example, it is configured to be received as a value of 75%. The battery maximum allowable current is a specification information of the battery pack, and a fixed value previously stored in the non-volatile memory of the battery management unit (230) is read. For example, it is configured to be read as a value of 100A.The final determination result of the above 'dynamic charging current control process' is the 'target charging current value' to be applied to the battery. The above 'target charging current value' is determined as a specific numerical value in amperes. This value is transmitted to the DC-DC converter controller of the power conversion unit (220). The above DC-DC converter performs optimal charging that matches the real-time state of the battery by precisely adjusting the output current to match the received 'target charging current value'. The scenario is explained as an example in which, while charging is in progress and the current charging state is 75%, 'Process 1' detects a gradual rise in battery temperature and issues a 'caution' state. First, the charging control module (250) receives data indicating a battery danger state 'caution', a current charging state of 75%, and a maximum allowable current of 100A. Second, since the received 'battery danger state' is 'caution', the upper control layer determines the safety limit current to 80A according to a predefined rule. Third, the lower control layer requests a maximum current of 100A because the current charging state is in the constant current range. Fourth, the lower value between the upper layer's determined value (80A) and the lower layer's requested value (100A) is selected as the final 'target charging current value' and determined to be 80A, and this value is transmitted to the power conversion unit (220) so that the actual charging current is immediately lowered to 80A.

[0270] Determination of Final Target Charging Current Based on Battery Condition (Example) (Maximum Allowable Current: 100A, CC-CV Switching SOC: 80%) scenario Input conditions judgment by hierarchy Final target current danger SOC Higher (Safety) Decision 1. Normal charging (initial) normal 30% No restrictions 2. Temperature rise detection caution 50% 80 A (20% decrease of 100A) 3. Normal Charging (Review) normal 85% No restrictions 4. Temperature rise + Review caution 85% 80 A 5. Severe temperature rise warning 60% 0 A (Charging stopped)

[0271] [Table 12] above is an example showing how the hierarchical control logic of the present invention determines the final target charging current according to various battery state combinations. In Scenario 2, when the danger state is determined to be 'Caution', the upper safety layer sets the safety limit current to 80A. The lower efficiency layer requests 100A because it is in the CC section, but the final target current is determined to be 80A, which is the lower of the two values, thereby ensuring safety.

[0272] In particular, when a combination of a temperature rise ('caution') and a situation near full charge (SOC 85%) occurs, such as in Scenario 4, the upper layer determines 80A and the lower layer determines 70A. The present invention demonstrates that precise control is possible to simultaneously mitigate two different types of battery stress factors by selecting 70A, which is the more conservative (lower) value among these, as the final target current. This is a unique effect of the present invention that is differentiated from prior art that uses only a single rule. The control logic includes a watchdog timer-based safety mechanism to ensure the integrity of the control signal. The charge control module (250) periodically transmits a control signal to the power conversion unit (220). If the power conversion unit (220) does not receive this periodic signal for a certain period of time, it is considered an error in the control system, and the charging current is immediately cut off to zero independently according to its own safety logic. Compared to prior art, in a scenario where the battery temperature rises gradually during charging, the prior art of a simple constant current-constant voltage control method continues to apply the maximum current regardless of signs of danger. On the other hand, the present invention detects a 'caution' state and preemptively lowers the current to 80A to induce temperature stabilization. In a scenario where a transient error occurs in the control system, the prior art may be exposed to the risk of overcharging, but the watchdog timer of the present invention detects the interruption of the control signal and ensures safety by having the power conversion unit independently cut off the current to zero. The 'safety-first hierarchical control' method of the present invention is an essential configuration for simultaneously achieving two potentially conflicting goals, 'efficiency' and 'safety,' in a system for charging high-output, high-capacity batteries. By adopting the safety-first hierarchical control rule as described above, the present invention achieves technical effects such as extending battery life, maximizing safety, improving charging reliability, and implementing intelligent charging.Specifically, it minimizes overload and thermal stress, includes specific response measures for exceptional situations, and dynamically adjusts the charging intensity according to the state of the battery. The criteria for the core threshold used in this specification are defined as follows. The constant current-constant voltage switching threshold is derived from the chemical characteristics of the lithium-ion battery used and is an objective switching point confirmed through the manufacturer's datasheet or charge-discharge cycle tests. The safety reduction ratio refers to the minimum current reduction range capable of effectively slowing down the rate of increase in thermal stress applied to the battery in response to specific risk signs, and is an engineering value set through thermal modeling simulations or iterative verification tests. Assuming a situation where heavy equipment finishes its work and enters a designated wireless charging zone, the entire operation process of a wireless charging-supported battery system for heavy equipment according to an embodiment of the present invention is described in detail. When the heavy equipment is positioned on the charging pad, the position alignment module (270) is activated. The position alignment module (270) analyzes the strength or pattern of the magnetic field detected by the receiving coil unit (210) and calculates the center axis alignment error with the transmitting coil unit (120) in real time. The alignment status is visually displayed on the driver's seat display, guiding the driver to fine-tune the vehicle to park in the optimal position. When the alignment error enters a preset allowable range, the position alignment module (270) transmits a 'alignment complete' signal to the central control unit (240). Upon receiving the 'alignment complete' signal, the central control unit (240) transmits a 'start charging' command to the power transmission unit (100) via the communication module (260). At the same time, it receives the current battery status from the battery management unit (230). The central control unit (240) inputs the received battery status into 'Process 1: Battery Risk Status Determination Process'.Since all values ​​are within the normal range, the final 'battery danger state' is determined to be 'normal'. Since it is in a 'normal' state, the central control unit (240) executes 'Process 4: Dynamic charging current control process'. The upper safety layer does not intervene, and the lower efficiency layer determines that it is a constant current charging section based on the current charging state, and determines the maximum allowable battery current as the 'target charging current value'. This value is transmitted to the power conversion unit (220), and normal charging begins. While charging is in progress, a small metal can blown by the wind rolls onto the charging pad and lands between the transmitting coil and the receiving coil. The central control unit (240) continuously performs 'Process 3: Charging abnormal state detection process'. Due to the metal foreign object, eddy current loss occurs, causing the 'actual charging power' measured by the power receiving unit (200) to decrease rapidly. The error rate between the 'expected charging power' calculated based on the current perfect alignment state and low temperature continuously exceeds the allowable threshold for more than 5 seconds. Accordingly, Process 3 ultimately determines the state to be 'charge abnormal'. The central control unit (240) transmits a command to the power transmitter (100) via the communication module (260) to stop charging immediately and sends a "Charging System Check: Foreign Substance or System Error Detected" warning to the driver. It assumes a worst-case scenario where the driver does not notice the warning and the foreign substance continues to overheat, applying thermal stress to the bottom of the battery. As a result, some battery cells begin to be damaged. The battery management unit (230) detects that the battery temperature has risen rapidly to 61°C and the gas sensor concentration to 210 ppm, and transmits this data to the central control unit (240) in real time. The central control unit (240) evaluates the new data through 'Process 1: Battery Risk State Determination Process'.Since both the temperature and gas exceed the threshold values, the final 'battery danger state' is immediately upgraded to 'emergency' according to the first rule of the state determination rule table. As soon as the 'emergency' state is determined, the central control unit (240) transmits an 'emergency detachment' command to the battery replacement unit (300). The control unit of the battery replacement unit (300) executes 'Process 2: Emergency Detachment Drive Control Process'. Based on parameters stored in memory, the optimal drive pressure required to safely detach the battery within 1.5 seconds is calculated. A drive signal corresponding to the calculated pressure is transmitted to the hydraulic system, causing multiple actuators to forcefully and rapidly push the battery pack out. The battery pack is released from the locking device and is physically and completely separated from the heavy equipment body along the designed path. All of these processes are completed within 1.5 seconds after determining an 'emergency' state, thereby fundamentally preventing the spread of fire to the entire vehicle. As such, in the embodiment of the present invention, each component and process organically exchanges data with one another, performing optimal judgment and control for each stage, ranging from normal charging to unpredictable abnormal states and critical danger situations. This goes beyond simple wireless charging functions and is a system that integrates intelligent state diagnosis and active safety response functions, thereby producing the effect of significantly improving the stability and reliability of heavy equipment operation.

[0274] The embodiments described above are for illustrative purposes only, and those skilled in the art will understand that the embodiments described above can be easily modified into other specific forms without altering the technical concept or essential features of the embodiments described above. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single unit may be implemented in a distributed manner, and components described as distributed may likewise be implemented in a combined form.

[0276] The scope of protection sought through this specification is defined by the claims set forth below rather than by the detailed description, and should be interpreted to include all modifications or variations derived from the meaning and scope of the claims and the concept of equivalents. Explanation of the symbols

[0277] 100: Management Server

Claims

Claim 1 A management server including a temperature sensing sensor; wherein the management server comprises: a temperature sensing module that detects and estimates the internal temperature of at least one battery in real time; a heating control module that controls a heating module provided on one side of the battery to drive the battery to heat when the temperature is low and release heat; and a cooling control module that controls a cooling module provided on one side of the battery to drive the battery to cool when the temperature is high and release cold air. The temperature correction module includes: generating a control command to control the operation of the heating module and the cooling module so as to maintain a preset appropriate temperature range of the battery, and transmitting the generated control command to the heating control module or the cooling control module; wherein the temperature sensing module estimates the internal temperature of the battery based on a plurality of temperature sensing sensors already provided on the exterior of the battery and the surface temperature and ambient temperature of the battery detected by the plurality of temperature sensing sensors, and estimates the internal temperature of the battery by reflecting the surface temperature and ambient temperature of the battery and the model information of the battery; wherein the temperature sensing module communicates in real time with the temperature correction module via a network, assigns a unique ID to each battery and stores it in a database, and simultaneously provides the unique ID of each battery, the current internal temperature of the battery, and the current time to the temperature correction module; wherein the temperature sensing module provides the unique ID of each battery, the current internal temperature of the battery, and the current time to the temperature correction module at predetermined intervals; and wherein the preset appropriate temperature range of the battery communicates with the management server. It refers to an area between the lowest temperature and the highest temperature input from a portable administrator terminal, and the temperature correction module generates a control command to operate the heating module and stop the operation of the cooling module when the current internal temperature of a specific battery is lower than the preset appropriate temperature range of the battery, and transmits the command to the heating control module and the cooling control module, and the temperature correction module,If the current internal temperature of a specific battery is higher than the preset appropriate temperature range of the battery, the cooling module is driven and the heating module is stopped. A control command is generated and transmitted to the heating control module and the cooling control module. The heating module refers to a module capable of generating heat to raise the internal temperature of the battery, including a PTC heater, a heat pad, and a Peltier element. The amount of heat generated by the heating module is continuously variablely controlled by a control command received from the heating control module. The cooling module refers to a module capable of releasing cold air to lower the internal temperature of the battery, including a fan, a pump, a valve, a liquid cooler, a heat sink, and a Peltier element. The amount of cooling generated by the cooling module is continuously variablely controlled by a control command received from the cooling control module. If there is a battery that falls outside the error range of the preset appropriate temperature range of the battery, the temperature correction module sends the unique ID of the battery, the current internal temperature of the battery, and the current time to the administrator terminal. The error range of the preset appropriate temperature range of the battery is determined by the administrator terminal. A battery management device for low and high temperature environments, referring to the input range. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 delete Claim 6 delete

Images