Blade battery temperature control method and system based on fuzzy neural network

Abstract

The invention belongs to the technical field of battery temperature control, and provides a blade battery temperature control method and system based on a fuzzy neural network, and the method comprises the following steps: monitoring the temperature of a battery cell and a phase change material in real time, monitoring the charging and discharging current of a battery in real time through a current transformer, and obtaining current data; preprocessing the battery cell temperature data, the phase change material temperature data and the current data to obtain preprocessed data; the preprocessed data serve as input variables and are input into a fuzzy neural network for fuzzy processing, and a temperature adjusting instruction is generated according to the preset working temperature range of the blade battery; according to the temperature adjusting instruction, the temperature of the blade battery is adjusted, and the temperature change is monitored in real time; through the complex mapping capability of the fuzzy neural network, accurate control over the battery temperature is achieved, temperature fluctuation is reduced, the consistency of battery performance is improved, and the temperature control performance of the blade battery is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery temperature control, and in particular to a blade battery temperature control method and system based on a fuzzy neural network. Background Art

[0002] With the rapid development and widespread application of battery technology, especially the increasing importance of blade batteries in electric vehicles and energy storage systems, battery temperature control technology has become increasingly critical. Temperature has a significant impact on battery performance, safety and service life;

[0003] However, common blade battery temperature control technologies mainly include traditional PID control, model-based control, and simple threshold control methods. Although traditional PID control shows good control effects in some linear systems, it has obvious limitations for systems with complex nonlinear characteristics such as blade batteries. Parameters such as the internal resistance and chemical reaction rate of the battery will change significantly with changes in temperature, charge and discharge rate, and battery aging. This makes it difficult for PID control to accurately establish a control model and accurately describe the complex nonlinear relationship between battery temperature and various influencing factors, resulting in low temperature control accuracy and large battery temperature fluctuations, which seriously affect the consistency of battery performance. For example, when charging at a high rate, the heat generated by the battery increases rapidly. Traditional PID control may not be able to adjust the heat dissipation power in time, causing the battery temperature to be too high, thereby reducing the battery's charge and discharge efficiency and even causing safety hazards;

[0004] In addition, the existing threshold control method starts the heating or cooling equipment when the battery temperature reaches a preset threshold. This method is too simple and crude, and lacks comprehensive consideration of the battery temperature change trend and other related factors. On the one hand, since the threshold setting is often fixed and cannot adapt to the needs of the battery under different working conditions, it is easy to start and stop the heating or cooling equipment frequently, which not only increases the energy consumption and wear of the equipment, but also may cause drastic fluctuations in the battery temperature. On the other hand, when the battery temperature hovers near the threshold, simple threshold control cannot provide fine-grained adjustment, making it difficult to ensure that the battery is always in the optimal operating temperature range, which is not conducive to extending battery life and optimizing performance;

[0005] To this end, those skilled in the art have proposed a blade battery temperature control method and system based on fuzzy neural networks, aiming to achieve more accurate, stable and reliable temperature control through complex mapping capabilities and real-time feedback mechanisms, thereby improving the temperature control performance of blade batteries. Summary of the Invention

[0006] In order to solve the above technical problems, the present invention provides a blade battery temperature control method and system based on a fuzzy neural network to solve the problems raised in the background technology.

[0007] According to the first aspect of the present disclosure, a blade battery temperature control method based on a fuzzy neural network is proposed, comprising the following steps:

[0008] S1. Use a thermistor sensor to sense temperature fluctuations through current changes, monitor the temperature of the battery cell and phase change material in real time, and obtain battery cell temperature data and phase change material temperature data; and use a current transformer to monitor the battery charge and discharge current in real time to obtain current data;

[0009] S2, converting the battery core temperature data, phase change material temperature data, and current data into electrical signals through a conditioning circuit, and transmitting the electrical signals to an intelligent temperature remote transmission detector for preprocessing, wherein the preprocessing includes cleaning, denoising, and normalization to obtain preprocessed data;

[0010] S3, inputting the pre-processed data as input variables into a fuzzy neural network for fuzzification processing, and generating a temperature adjustment instruction according to a preset operating temperature range of the blade battery;

[0011] S4. According to the temperature adjustment instruction, the blade battery is temperature-regulated, and the temperature change is monitored in real time. The new temperature data is fed back to the fuzzy neural network to form a closed-loop control.

[0012] Preferably, the pre-processed data is used as an input variable and input into a fuzzy neural network for fuzzification processing, and a temperature adjustment instruction is generated according to a preset operating temperature range of the blade battery, including:

[0013] By calculating the temperature data of the battery cell before and after the moment, the temperature change rate ΔT is obtained;

[0014] The pre-processed core temperature data, phase change material temperature data and current data are combined with the temperature change rate ΔT to form an input variable set X=[T a ,T b ,ΔT,I], where T a is the pre-processed battery cell temperature data, T b is the phase change material temperature data after preprocessing, and I is the current data after preprocessing;

[0015] Inputting the input variable set into a fuzzy neural network for fuzzification processing to obtain a heating instruction intensity and a cooling instruction intensity; the fuzzification processing includes: obtaining a fuzzification result through a fuzzification process; obtaining fuzzy membership function values ​​of the heating instruction intensity and the cooling instruction intensity through a fuzzy inference process; and obtaining an accurate heating instruction intensity and an accurate cooling instruction intensity through a defuzzification process;

[0016] According to the preset operating temperature range of the blade battery (T min,T max ), combining the heating instruction strength and the heat dissipation instruction strength to generate the final temperature adjustment instruction.

[0017] Preferably, obtaining a fuzzy result through a fuzzification process includes:

[0018] The pre-processed core temperature data T is calculated using the following formula: a , Phase change material temperature data T b , current data I and temperature change rate ΔT for their respective fuzzy subsets:

[0019]

[0020] Among them, x i is the i-th variable in the input variable set X, a i1 is the lower limit of the range of the i-th variable, a i2 is the upper limit of the range of the i-th variable, and a i1 <a i2 ,μ(x i ) is the degree of membership;

[0021] Each element in the input variable set X is mapped to the corresponding fuzzy subset membership value, and the fuzzified vector μ=[μ1,μ2,...,μ n ], where n is the total number of fuzzy subsets, each μ i Indicates the degree to which the input variable belongs to the corresponding fuzzy subset, the fuzzy vector μ=[μ1,μ2,...,μ n ] as the fuzzification result.

[0022] Preferably, the fuzzy membership function values ​​of the heating instruction intensity and the heat dissipation instruction intensity are obtained through the fuzzy inference process, including:

[0023] According to the fuzzification result μ=[μ1,μ2,...,μ n ], combined with the kth rule in the set fuzzy rule base, the antecedent satisfaction degree α is obtained k =min(μ(x1),μ(x2),...,μ(x n ));

[0024] Based on the degree of satisfaction of the antecedents and in combination with the kth rule in the fuzzy rule base, the fuzzy membership function value of the heating instruction intensity is obtained as follows: Among them H f is the fuzzy subset of the heating instruction intensity in the rule consequent, is the fuzzy subset membership function of heating instruction intensity;

[0025] Based on the degree of satisfaction of the antecedents and in combination with the kth rule in the fuzzy rule base, the fuzzy membership function value of the heat dissipation instruction intensity is obtained as follows: Among them C f is the fuzzy subset of the heat dissipation instruction intensity in the rule consequence, is the fuzzy subset membership function of the heat dissipation instruction intensity;

[0026] According to the fuzzy subset membership function of the heating instruction intensity, the fuzzy membership function value of the heating instruction intensity is According to the fuzzy subset membership function of the heat dissipation instruction intensity, the fuzzy membership function value of the heat dissipation instruction intensity is obtained as follows:

[0027] Preferably, obtaining the precise heating instruction intensity and the precise cooling instruction intensity through the defuzzification process includes:

[0028] According to the fuzzy membership function value of the heating instruction intensity, the center of gravity method is used to convert it into an accurate value, and the obtained accurate heating instruction intensity is:

[0029] According to the fuzzy membership function value of the heat dissipation instruction intensity, it is converted into an accurate value through the centroid method, and the accurate heat dissipation instruction intensity is obtained as follows:

[0030] Among them, H i 、C i are discrete values ​​in the domain of heating instruction intensity and cooling instruction intensity, μ H (H i ), μ C (C i ) are the membership values ​​of the corresponding discrete values, N and M are the numbers of discrete values ​​in the domain of heating instruction intensity and cooling instruction intensity respectively;

[0031] According to the precise heating instruction intensity H0 and the precise heat dissipation instruction intensity C0, and the preset operating temperature range (T min ,T max ) to generate the final temperature adjustment instruction.

[0032] According to the second aspect of the present disclosure, a blade battery temperature control system based on a fuzzy neural network is also proposed, comprising:

[0033] The data acquisition module is used to monitor the temperature of the battery cells and phase change materials in real time to obtain the battery cell temperature data and the phase change material temperature data; and uses the current transformer to monitor the battery charge and discharge current in real time to obtain the current data;

[0034] The data preprocessing module is used to convert the battery core temperature data, phase change material temperature data and current data into electrical signals through the conditioning circuit, and transmit them to the intelligent temperature remote detection instrument for preprocessing to obtain the preprocessed data.

[0035] A fuzzy neural network control module is used to input the pre-processed data as an input variable into a fuzzy neural network for fuzzification processing, and generate a temperature adjustment instruction according to a preset operating temperature range of the blade battery;

[0036] An execution module, configured to adjust the temperature of the blade battery according to the temperature adjustment instruction, start a heating or cooling device, and adjust the temperature;

[0037] The feedback module is used to monitor temperature changes in real time, feed back new temperature data to the fuzzy neural network, re-fuzzify it, generate new temperature adjustment instructions, and continuously cycle to form a closed-loop control.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] 1. The present invention combines fuzzy logic and neural networks, utilizing both fuzzy logic to process uncertain information and the self-learning and adaptive characteristics of neural networks. This overcomes the problem of inaccurate description of complex nonlinear characteristics of batteries by traditional control methods. Through the complex mapping capability of fuzzy neural networks, it can more accurately process the nonlinear relationship between battery temperature and various influencing factors, achieve precise control of battery temperature, reduce temperature fluctuations, and improve the consistency of battery performance.

[0040] 2. The present invention uses fuzzy processing and rule-based reasoning to make the system more tolerant to data noise and uncertainty, and can maintain good control effects under different environmental conditions and battery aging conditions, thereby improving the stability and reliability of the system.

[0041] 3. The present invention forms a closed-loop control through real-time monitoring and feedback, and can adjust the control strategy in time according to the dynamic changes in battery temperature. It has higher control accuracy and dynamic response capability, and better meets the strict temperature control requirements of blade batteries in different usage scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a flow chart of the blade battery temperature control method based on fuzzy neural network of the present invention;

[0043] Figure 2 This is a block diagram of the blade battery temperature control system based on fuzzy neural network of the present invention. DETAILED DESCRIPTION

[0044] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0045] As attached Figure 1 As shown:

[0046] Embodiment 1: The present invention provides a blade battery temperature control method based on a fuzzy neural network, comprising the following steps:

[0047] S1. Use thermistor sensors to sense temperature fluctuations through current changes, monitor the temperature of battery cells and phase change materials in real time, and obtain battery cell temperature data and phase change material temperature data; and use current transformers to monitor battery charge and discharge current in real time to obtain current data; add redundant sensors or introduce multiple types of sensors on the basis of thermistor sensors to cross-validate data, thereby avoiding the risk of single point failure.

[0048] S2, converting the battery core temperature data, phase change material temperature data, and current data into electrical signals through a conditioning circuit, and transmitting the signals to an intelligent temperature remote transmission detector for preprocessing to obtain preprocessed data;

[0049] S3. Inputting the preprocessed data as an input variable into a fuzzy neural network for fuzzification processing, and generating a temperature adjustment instruction according to a preset operating temperature range of the blade battery; Inputting the preprocessed data as an input variable into a fuzzy neural network for fuzzification processing, and generating a temperature adjustment instruction according to a preset operating temperature range of the blade battery, including:

[0050] By calculating the temperature data of the battery cell before and after, the temperature change rate ΔT is obtained;

[0051] The pre-processed core temperature data, phase change material temperature data and current data are combined with the temperature change rate ΔT to form the input variable set X = [T a ,T b ,ΔT,I], where T a is the pre-processed battery cell temperature data, T b is the phase change material temperature data after preprocessing, and I is the current data after preprocessing;

[0052] Inputting the input variable set into the fuzzy neural network for fuzzification processing to obtain the heating instruction intensity and the cooling instruction intensity; the fuzzification processing includes the fuzzification process, the fuzzy reasoning process and the defuzzification process;

[0053] The intensity of the heating instruction and the intensity of the heat dissipation instruction are determined according to the preset operating temperature range (T min ,T max ) to generate the final temperature adjustment instruction.

[0054] The fuzzification process includes:

[0055] Calculate the pre-processed cell temperature data T using the following formula: a , Phase change material temperature data T b , current data I and temperature change rate ΔT for their respective fuzzy subsets:

[0056]

[0057] Among them, x i is the i-th variable in the input variable set X, a i1 is the lower limit of the range of the i-th variable, a i2 is the upper limit of the range of the i-th variable, and a i1 <a i2 ,μ(x i ) is the degree of membership;

[0058] Map each element in the input variable set X to the corresponding fuzzy subset membership value, and get the fuzzy vector μ=[μ1,μ2,...,μ n ], where n is the total number of fuzzy subsets, each μ i Indicates the degree to which the input variable belongs to the corresponding fuzzy subset. The fuzzified vector μ=[μ1,μ2,...,μ n ] as the fuzzification result.

[0059] The collected precise data is converted into fuzzy sets, and the numerical values ​​are mapped to corresponding fuzzy linguistic variables by defining membership functions, so that uncertain and imprecise information can be processed, which is more in line with the complex situations in the actual operation of the battery, enhances the adaptability to different working conditions, and avoids overly strict threshold judgments that may occur in precise control.

[0060] The fuzzy reasoning process includes:

[0061] According to the fuzzification result μ=[μ1,μ2,...,μ n ], combined with the kth rule in the set fuzzy rule base, the antecedent satisfaction degree α is obtained k =min(μ(x1),μ(x2),...,μ(x n ));

[0062] The fuzzy membership function value of the heating instruction intensity corresponding to the kth rule in the fuzzy rule base is: Among them H f is the fuzzy subset of the heating instruction intensity in the rule consequent, is the fuzzy subset membership function of heating instruction intensity;

[0063] The fuzzy membership function value of the heat dissipation instruction intensity corresponding to the kth rule in the fuzzy rule base is: Among them C f is the fuzzy subset of the heat dissipation instruction intensity in the rule consequence, is the fuzzy subset membership function of the heat dissipation instruction intensity;

[0064] The fuzzy membership function value of the heating instruction intensity is obtained as The fuzzy membership function value of the heat dissipation instruction intensity is

[0065] Drawing on the expertise and experience of battery experts, a series of fuzzy rules were developed to describe the relationship between input variables and output variables (such as heating or cooling instructions). This provides a basis for fuzzy reasoning, enabling reasonable temperature adjustment decisions based on different battery states. Based on the fuzzified input and the fuzzy rule base, a fuzzy reasoning algorithm is applied to derive the fuzzy output—the fuzzy decision on battery temperature adjustment. By comprehensively considering the influence of multiple input variables through the fuzzy reasoning process, more accurate temperature adjustment decisions can be made under complex battery operating conditions, avoiding the limitations of single-variable control and improving the accuracy and rationality of temperature control.

[0066] The defuzzification process includes:

[0067] The fuzzy membership function values ​​of the heating instruction intensity and the fuzzy membership function values ​​of the cooling instruction intensity are converted into precise values ​​by using the center of gravity method;

[0068] The precise heating instruction intensity is obtained as

[0069] The precise heat dissipation instruction intensity is obtained as

[0070] Among them, H i 、C i are discrete values ​​in the domain of heating instruction intensity and cooling instruction intensity, μ H (H i ), μ C (C i ) are the membership values ​​of the corresponding discrete values, N and M are the numbers of discrete values ​​in the domain of heating instruction intensity and cooling instruction intensity respectively;

[0071] According to the precise heating instruction intensity H0 and the precise heat dissipation instruction intensity C0, and the preset operating temperature range of the blade battery (T min ,T max) to generate the final temperature adjustment command. The fuzzy output is converted into a precise control variable, such as the specific heating power or cooling fan speed, so that the actual actuator can adjust the battery temperature based on this control variable. By converting fuzzy decisions into precise, executable control signals, the transition from fuzzy logic to actual physical control is achieved, making temperature adjustment operational.

[0072] S4. According to the temperature adjustment instruction, the blade battery is temperature-regulated, and the temperature change is monitored in real time. The new temperature data is fed back to the fuzzy neural network to form a closed-loop control;

[0073] When the heating instruction intensity H0 generated by the fuzzy neural network is greater than the set threshold and the current battery temperature T current Lower than the preset operating temperature range lower limit T min When the heating device is started, the battery temperature change is monitored in real time during the heating process. After a time Δt, the battery temperature change Δt heat The actual measured battery temperature T is estimated by the heat transfer equation. current Continuously update to get new temperature data, and the new temperature data T new The data is collected by thermistor sensor and transmitted to the fuzzy neural network;

[0074] When the heat dissipation instruction intensity C0 is greater than the set threshold and the current battery temperature T current Higher than the preset upper limit of the operating temperature range T max , start the heat dissipation device; during the heat dissipation process, monitor the battery temperature changes in real time to obtain new temperature data, and the new temperature data T new The data is collected by thermistor sensor and transmitted to the fuzzy neural network;

[0075] The fuzzy neural network is based on the new feedback temperature data T new , combined with the current data I, the fuzzification, fuzzy reasoning and defuzzification process are re-performed to generate new temperature adjustment instructions. The whole process is continuously cycled to form a closed-loop control, so that the battery temperature is always maintained within the preset operating temperature range.

[0076] According to the control quantity obtained by defuzzification, the heating or cooling equipment is driven to adjust the temperature of the blade battery, and the temperature changes are monitored in real time. The new temperature data is fed back to the fuzzy neural network to form a closed-loop control, so that the battery temperature can be quickly and accurately stabilized within the preset operating temperature range, thereby improving the safety and service life of the battery. At the same time, the closed-loop control can respond to changes in the battery status in a timely manner and dynamically adjust the temperature, ensuring the stability and reliability of temperature control.

[0077] By forming a closed-loop control through real-time monitoring and feedback, the control strategy can be adjusted in time according to the dynamic changes in battery temperature. Compared with some open-loop control or simple feedback control methods, it has higher control accuracy and dynamic response capabilities, and better meets the strict temperature control requirements of blade batteries in different usage scenarios.

[0078] As attached Figure 2 As shown:

[0079] Embodiment 2: The present invention provides a blade battery temperature control system based on a fuzzy neural network, comprising:

[0080] The data acquisition module is used to monitor the temperature of the battery cells and phase change materials in real time to obtain the battery cell temperature data and the phase change material temperature data; and uses the current transformer to monitor the battery charge and discharge current in real time to obtain the current data;

[0081] The data preprocessing module is used to convert the battery core temperature data, phase change material temperature data and current data into electrical signals through the conditioning circuit, and then transmit them to the intelligent temperature remote detection instrument for preprocessing to obtain the preprocessed data.

[0082] The fuzzy neural network control module is used to input the pre-processed data as input variables into the fuzzy neural network for fuzzification processing, and generate temperature adjustment instructions according to the preset operating temperature range of the blade battery;

[0083] An execution module is used to adjust the temperature of the blade battery according to the temperature adjustment instruction, start the heating or cooling device, and adjust the temperature;

[0084] The feedback module is used to monitor temperature changes in real time, feed back new temperature data to the fuzzy neural network, re-fuzzify it, generate new temperature adjustment instructions, and continuously cycle to form a closed-loop control.

[0085] From the above, we can see that through the complex mapping capabilities of fuzzy neural networks, the nonlinear relationship between battery temperature and various influencing factors can be processed more accurately, the battery temperature can be precisely controlled, temperature fluctuations can be reduced, and the consistency of battery performance can be improved; fuzzy processing and rule-based reasoning methods enable the system to have a strong tolerance for data noise and uncertainty, and can maintain good control effects under different environmental conditions and battery aging conditions, thereby improving the stability and reliability of the system; precise temperature control helps to keep the battery within a suitable operating temperature range, reduce the heterogeneity of chemical reactions inside the battery, reduce the battery aging rate, extend the battery life, and at the same time improve the battery's charge and discharge efficiency and optimize the battery's overall performance.

[0086] Example 3: Based on the blade battery temperature control method provided in Example 1, an online learning mechanism is introduced to dynamically adjust the neural network weights and fuzzy rules in combination with real-time data to improve the long-term adaptability of the system; that is, the model is trained using historical data to achieve adaptive temperature control in the entire temperature range.

[0087] In addition, in addition to the input variables: battery cell temperature data, phase change material temperature data, current data and temperature change rate, potential influencing factors such as ambient humidity and battery state of charge are added as supplementary multi-source data input, thereby increasing the model's ability to analyze complex working conditions.

[0088] Experimental example: In extreme cold, the innovative PTC dual-circulation preheating technology relies on a three-channel heating module and symmetrical pipes to achieve 40% efficient rapid equalization of battery heat at -30°C. A two-way valve intelligently switches the coolant flow direction, directing it to the heat dissipation module for forced cooling at high temperatures and connecting it to the PTC for precise heating at low temperatures. This creates a full-range temperature control closed loop covering -30°C to 60°C, ultimately controlling battery temperature fluctuations within ±2°C and improving overall energy efficiency by 25%. At the same time, with the help of fuzzy control algorithms and mature neural networks, the optimal PTC power is optimized during heating, and the optimal coolant flow rate is determined during heat dissipation, significantly improving temperature control efficiency, reducing energy consumption, and extending battery life.

[0089] It is important to note that the construction and arrangement of the present application shown in a plurality of different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, it will be readily understood by those who consult this disclosure that many modifications are possible without departing substantially from the novel teachings and advantages of the subject matter described in this application. Other replacements, modifications, changes, and omissions may be made in the design, operating conditions, and arrangement of the exemplary embodiments without departing from the scope of the present invention. Therefore, the present invention is not limited to specific embodiments, but extends to a variety of modifications still falling within the scope of the appended claims.

[0090] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment (i.e., those features that are not relevant to the best mode presently contemplated for carrying out the invention or those that are not relevant to implementing the invention) may not be described.

[0091] It will be understood that in the development of any actual embodiment, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but for those of ordinary skill having the benefit of this disclosure, the development effort will be a routine task of design, fabrication, and production without undue experimentation.

[0092] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A blade battery temperature control method based on fuzzy neural network, characterized in that: The following steps are involved: S1. Use a thermistor sensor to sense temperature fluctuations through current changes, monitor the temperature of the battery cell and phase change material in real time, and obtain battery cell temperature data and phase change material temperature data; and use a current transformer to monitor the battery charge and discharge current in real time to obtain current data; S2, converting the battery core temperature data, phase change material temperature data, and current data into electrical signals through a conditioning circuit, and transmitting the electrical signals to an intelligent temperature remote transmission detector for preprocessing, wherein the preprocessing includes cleaning, denoising, and normalization to obtain preprocessed data; S3, inputting the pre-processed data as input variables into a fuzzy neural network for fuzzification processing, and generating a temperature adjustment instruction according to a preset operating temperature range of the blade battery; S4. According to the temperature adjustment instruction, the blade battery is temperature-adjusted, and the temperature change is monitored in real time. The new temperature data is fed back to the fuzzy neural network to form a closed-loop control.

2. The blade battery temperature control method based on fuzzy neural network according to claim 1, characterized in that: The pre-processed data is used as an input variable and input into a fuzzy neural network for fuzzification processing, and a temperature adjustment instruction is generated according to a preset operating temperature range of the blade battery, including: By calculating the temperature data of the battery cell before and after the moment, the temperature change rate ΔT is obtained; The pre-processed core temperature data, phase change material temperature data and current data are combined with the temperature change rate ΔT to form an input variable set X=[T a ,T b ,ΔT,I], where T a is the pre-processed battery cell temperature data, T b is the phase change material temperature data after preprocessing, and I is the current data after preprocessing; Inputting the input variable set into a fuzzy neural network for fuzzification processing to obtain a heating instruction intensity and a cooling instruction intensity; the fuzzification processing includes: obtaining a fuzzification result through a fuzzification process; obtaining fuzzy membership function values ​​of the heating instruction intensity and the cooling instruction intensity through a fuzzy inference process; and obtaining an accurate heating instruction intensity and an accurate cooling instruction intensity through a defuzzification process; According to the preset operating temperature range of the blade battery (T min ,T max ), combining the heating instruction strength and the heat dissipation instruction strength to generate the final temperature adjustment instruction.

3. The blade battery temperature control method based on fuzzy neural network as claimed in claim 2, characterized in that: The fuzzification result is obtained through the fuzzification process, including: The pre-processed core temperature data T is calculated using the following formula: a , Phase change material temperature data T b , current data I and temperature change rate ΔT for their respective fuzzy subsets: Among them, x i is the i-th variable in the input variable set X, a i1 is the lower limit of the range of the i-th variable, a i2 is the upper limit of the range of the i-th variable, and a i1 <a i2 ,μ(x i ) is the degree of membership; Each element in the input variable set X is mapped to the corresponding fuzzy subset membership value, and the fuzzified vector μ=[μ1,μ2,...,μ n ], where n is the total number of fuzzy subsets, each μ i Indicates the degree to which the input variable belongs to the corresponding fuzzy subset, the fuzzy vector μ=[μ1,μ2,...,μ n ] as the fuzzification result.

4. The blade battery temperature control method based on fuzzy neural network as claimed in claim 3, characterized in that: The fuzzy membership function values ​​of the heating instruction intensity and the heat dissipation instruction intensity are obtained through the fuzzy reasoning process, including: According to the fuzzification result μ=[μ1,μ2,...,μ n ], combined with the kth rule in the set fuzzy rule base, the antecedent satisfaction degree α is obtained k =min(μ(x1),μ(x2),...,μ(x n )); Based on the degree of satisfaction of the antecedents and in combination with the kth rule in the fuzzy rule base, the fuzzy membership function value of the heating instruction intensity is obtained as follows: Among them H f is the fuzzy subset of the heating instruction intensity in the rule consequent, is the fuzzy subset membership function of heating instruction intensity; Based on the degree of satisfaction of the antecedents and in combination with the kth rule in the fuzzy rule base, the fuzzy membership function value of the heat dissipation instruction intensity is obtained as follows: Among them C f is the fuzzy subset of the heat dissipation instruction intensity in the rule consequence, is the fuzzy subset membership function of the heat dissipation instruction intensity; According to the fuzzy subset membership function of the heating instruction intensity, the fuzzy membership function value of the heating instruction intensity is According to the fuzzy subset membership function of the heat dissipation instruction intensity, the fuzzy membership function value of the heat dissipation instruction intensity is obtained as follows:

5. The blade battery temperature control method based on fuzzy neural network as claimed in claim 4, characterized in that: The accurate heating instruction intensity and the accurate cooling instruction intensity are obtained through the defuzzification process, including: According to the fuzzy membership function value of the heating instruction intensity, the center of gravity method is used to convert it into an accurate value, and the obtained accurate heating instruction intensity is: According to the fuzzy membership function value of the heat dissipation instruction intensity, it is converted into an accurate value through the centroid method, and the accurate heat dissipation instruction intensity is obtained as follows: Among them, H i 、C i are discrete values ​​in the domain of heating instruction intensity and cooling instruction intensity, μ H (H i ), μ C (C i ) are the membership values ​​of the corresponding discrete values, N and M are the numbers of discrete values ​​in the domain of heating instruction intensity and cooling instruction intensity respectively; According to the precise heating instruction intensity H0 and the precise heat dissipation instruction intensity C0, and the preset operating temperature range (T min ,T max ) to generate the final temperature adjustment instruction.

6. The blade battery temperature control system based on fuzzy neural network is characterized by: include: The data acquisition module is used to monitor the temperature of the battery cells and phase change materials in real time and obtain the battery cell temperature data and the phase change material temperature data; And use the current transformer to monitor the battery charging and discharging current in real time to obtain current data; The data preprocessing module is used to convert the battery core temperature data, phase change material temperature data and current data into electrical signals through the conditioning circuit, and transmit them to the intelligent temperature remote detection instrument for preprocessing to obtain the preprocessed data. A fuzzy neural network control module is used to input the pre-processed data as an input variable into a fuzzy neural network for fuzzification processing, and generate a temperature adjustment instruction according to a preset operating temperature range of the blade battery; An execution module, configured to adjust the temperature of the blade battery according to the temperature adjustment instruction, start a heating or cooling device, and adjust the temperature; The feedback module is used to monitor temperature changes in real time, feed back new temperature data to the fuzzy neural network, re-fuzzify it, generate new temperature adjustment instructions, and continuously cycle to form a closed-loop control.

Images