CN-121983711-A - Dynamic regulation and control method for heat dissipation efficiency of power plant energy storage system

CN121983711ACN 121983711 ACN121983711 ACN 121983711ACN-121983711-A

Abstract

The invention provides a dynamic regulation and control method for heat dissipation efficiency of an energy storage system of a power plant, which belongs to the technical field of energy storage systems, and is characterized in that a hybrid neural network model is built based on combination of mechanism modeling and a data driving method to learn a nonlinear relation between heat generation power and temperature of an energy storage battery on line, a multi-sensor fusion technology is adopted to collect operation parameters of the energy storage system and calculate thermodynamic characteristic parameters to build a thermal state characteristic vector, a standard heat dissipation mode, a modified heat dissipation mode or an abnormal heat dissipation mode is adopted according to a similarity comparison result of the thermal state characteristic vector and the standard heat balance characteristic vector, and when the temperature change rate of a battery monomer exceeds a threshold value and the duration accords with a condition, quick response regulation or slow response regulation is executed, so that the technical problem of insufficient temperature control precision caused by response lag of the heat dissipation system of the energy storage battery is solved.

Inventors

WANG ZEZHONG
XIE JINCHAO
XU MEI
ZHANG CHENXI
ZHU CHANG
WEI FEI
BAI DINGRONG

Assignees

鄂尔多斯实验室
清华大学

Dates

Publication Date: 20260505
Application Date: 20251128

Claims (10)

1. A dynamic regulation and control method for heat dissipation efficiency of an energy storage system of a power plant is characterized in that a plurality of temperature sensors and heat flow sensors are uniformly arranged on the surface of an energy storage battery cell, a millisecond-level electrochemical reaction heat monitoring network is established, battery cell temperature distribution and heat generation power density are collected in real time, a multi-time-scale layered control architecture is constructed, a fast time-scale control layer is arranged for processing millisecond-level transient heat response, a slow time-scale control layer is used for optimizing an hour-level long-term heat management strategy, a hybrid neural network model is constructed based on combination of mechanism modeling and a data driving method, a nonlinear relation between heat generation power of the energy storage battery and temperature and heat dissipation system heat transfer characteristics are established, a multi-sensor fusion technology is adopted to collect operation parameters of the energy storage system, including battery cell voltage, battery cell current, battery cell temperature, heat dissipation fan rotating speed, cooling liquid flow and environmental temperature, thermodynamic characteristic parameters are calculated according to the battery cell voltage, battery cell temperature and battery cell heat generation power density, a temperature gradient change rate, a heat dissipation efficiency coefficient and a heat balance bias value are set up, a thermal state characteristic vector is established when the similarity of the thermal state characteristic vector and a standard thermal balance characteristic vector is larger than a first similarity, a second similarity threshold is adopted, when the similarity is higher than a first similarity threshold value and a second similarity threshold value is adopted, and a second similarity threshold value is adopted when the similarity threshold value is higher than a similarity threshold value is reached, and a similarity threshold value is immediately when a thermal threshold value is different than a threshold value is different, and a temperature is different, and if the temperature change rate of the battery cells exceeds the temperature change threshold and the duration is less than the duration threshold, executing slow response adjustment.
2. The method for dynamically adjusting and controlling the heat dissipation efficiency of an energy storage system of a power plant according to claim 1, wherein the multi-time-scale layered control architecture is characterized in that heat dissipation control of the energy storage system is divided into different layers according to time scales, response time of a fast time scale control layer is in millisecond level, transient temperature change caused by electrochemical reaction is mainly processed, response time of a slow time scale control layer is in minute-hour level, and overall thermal management strategy and energy efficiency ratio are mainly optimized.
3. The dynamic regulation and control method for the heat dissipation efficiency of the power plant energy storage system according to claim 2, wherein the hybrid neural network model is specifically combined with physical mechanism modeling and data driving learning, temperature distribution spatial features are extracted through a convolutional neural network, time sequence thermodynamic rules are learned through a long-term and short-term memory network, and accurate modeling and prediction of the nonlinear thermal characteristics of the energy storage system are achieved.
4. The method for dynamically controlling the heat dissipation efficiency of an energy storage system of a power plant according to claim 3, wherein the heat generation power density of the battery unit refers to the heat generated by a unit volume of the battery in a unit time, and is calculated by the voltage, the current and the efficiency loss of the battery unit, and is used for evaluating the heat generation intensity and the heat load distribution of the battery unit, and performing physical mechanism analysis according to an electrochemical reaction heat generation equation.
5. The method according to claim 4, wherein the equation of heat generation of electrochemical reaction is specifically used for describing a mechanism relationship of heat generation during electrochemical reaction in the battery, the input includes activation energy of electrode reaction, mobility of electrolyte ions, and rate constant of reaction on the surface of the electrode, and the output is rate of heat generation of electrochemical reaction.
6. The method for dynamically controlling the heat dissipation efficiency of an energy storage system of a power plant according to claim 5, wherein the temperature gradient change rate refers to a change speed of temperature distribution in a battery cell with time, and is obtained by differential calculation of temperatures of the battery cells measured by adjacent temperature sensors, and is used for identifying hot spot formation and temperature non-uniformity degree, and analyzing according to a thermal conduction dynamics equation.
7. The method according to claim 6, wherein the thermal conduction dynamics equation is specifically used for describing a physical mechanism process of heat transfer inside the battery, the input includes material thermal conductivity, temperature gradient, material density and specific heat capacity, and the output is thermal conduction rate.
8. The method for dynamically adjusting and controlling the heat dissipation efficiency of an energy storage system of a power plant according to claim 7, wherein the heat dissipation efficiency coefficient is specifically calculated by calculating a ratio of heat dissipation power to input power, the heat dissipation power is determined according to a rotation speed of a heat dissipation fan and a flow rate of cooling liquid, the input power is determined according to a voltage of a battery cell and a current of the battery cell, and the heat dissipation efficiency coefficient is evaluated by adopting a heat dissipation efficiency function.
9. The method according to claim 8, wherein the heat dissipation efficiency function is specifically used for quantifying a ratio of an actual heat dissipation capacity to a theoretical heat dissipation capacity of the heat dissipation system, the input includes a heat dissipation fan rotation speed, a coolant flow rate, and an ambient temperature, and the output is a heat dissipation efficiency coefficient.
10. The method for dynamically adjusting and controlling the heat dissipation efficiency of an energy storage system of a power plant according to claim 9, wherein the heat balance deviation value is obtained by calculating a difference between heat generation power and heat dissipation power, the heat generation power is determined according to the heat generation power density of the battery cell, and the heat balance deviation function is adopted for calculation.

Description

Dynamic regulation and control method for heat dissipation efficiency of power plant energy storage system Technical Field The invention belongs to the technical field of energy storage systems, and particularly relates to a dynamic regulation and control method for heat dissipation efficiency of an energy storage system of a power plant. Background The heat dissipation control of the traditional energy storage system mainly adopts a single time scale control method based on temperature feedback, and the heat balance of the system is maintained by monitoring the temperature change of the battery and adjusting the rotating speed of a fan or the flow of cooling liquid. In the operation of the current energy storage power station, as the electrochemical reaction of the battery has the characteristic of millisecond-level rapid change, the response time of the traditional heat dissipation control system is usually second-level or minute-level, and the obvious problem of mismatching of time scales exists, so that the system cannot timely respond to the rapid temperature change in the battery, and particularly under the high-power charge and discharge working condition, the phenomenon of nonuniform temperature distribution in the battery is more serious. That is, the prior art has the technical problem that the temperature control precision is insufficient due to the response lag of the heat dissipation system of the energy storage battery. Disclosure of Invention In view of the above, the invention provides a dynamic regulation and control method for heat dissipation efficiency of an energy storage system of a power plant, which can solve the technical problem of insufficient temperature control precision caused by response lag of the energy storage battery heat dissipation system in the prior art. The invention provides a dynamic regulation and control method for heat dissipation efficiency of a power plant energy storage system, which is realized by uniformly arranging a plurality of temperature sensors and heat flow sensors on the surface of an energy storage battery monomer, establishing a millisecond-level electrochemical reaction heat monitoring network, collecting battery monomer temperature distribution and heat generation power density in real time, constructing a multi-time-scale layered control architecture, setting a rapid time-scale control layer for processing millisecond-level transient heat response, setting a slow time-scale control layer for optimizing an hour-level long-term heat management strategy, combining a mechanism modeling and a data driving method, constructing a hybrid neural network model, learning the nonlinear relation between the heat generation power of the energy storage battery and the temperature and the heat dissipation system heat transfer characteristic on line, acquiring the operation parameters of the energy storage system by adopting a multi-sensor fusion technology, including battery monomer voltage, battery monomer current, battery monomer temperature, heat dissipation fan rotating speed, cooling liquid flow and environmental temperature, calculating thermodynamic characteristic parameters according to the battery monomer voltage, battery monomer current and battery monomer temperature, including battery monomer heat generation power density, temperature gradient change rate, heat dissipation efficiency coefficient and heat balance bias value, establishing a thermal state characteristic vector, when the thermal state characteristic vector and a standard heat balance characteristic vector are large in similarity to a first similarity, constructing a second similarity mode when the thermal state characteristic vector is similar to a first similarity threshold is larger than a second similarity mode, and a second similarity mode is used for immediately correcting the thermal threshold is similar to a threshold value when the first similarity is higher than a threshold value and a second similarity threshold value is different when a threshold value is different than a threshold value is different in a threshold value, and if the temperature change rate of the battery cells exceeds the temperature change threshold and the duration is less than the duration threshold, executing slow response adjustment. The multi-time-scale layered control architecture is characterized in that heat dissipation control of an energy storage system is divided into different layers according to time scales, response time of a fast time-scale control layer is in a millisecond level, transient temperature change caused by electrochemical reaction is mainly processed, response time of a slow time-scale control layer is in a minute-hour level, and overall thermal management strategy and energy efficiency ratio are mainly optimized. The hybrid neural network model is specifically combined with physical mechanism modeling and data driving learning, temperature distribution space characteristics are e