Temperature management method and system of liquid cooling string PCS energy storage converter

Abstract

The invention provides a temperature management method and system for a liquid cooling string PCS energy storage converter, and the method comprises the steps: carrying out the temperature data collection of each string in the liquid cooling string PCS energy storage converter, calculating the temperature characteristic parameters, and obtaining a string temperature state matrix; dividing cooling priorities of the strings according to the string temperature state matrix and carrying out cooling resource allocation to obtain a dynamic cooling execution instruction set; based on the dynamic cooling execution instruction set and the string temperature state matrix, identifying a hot point string and a cold point string, and calculating a power adjustment scheme to obtain a power distribution execution matrix; and selecting a corresponding temperature control strategy according to the environment state parameters and the power distribution execution matrix, and performing parameter setting on each component of the cooling system. According to the method, dynamic cooling resource allocation, power dynamic balance and environment adaptability temperature control strategies are combined, the problems of unreasonable cooling resource allocation and low temperature management efficiency are solved, and accurate temperature management of the liquid cooling string PCS energy storage converter under various working conditions is realized.

Description

Technical Field

[0001] The present invention relates to the field of simulation technology, and in particular to a temperature management method and system for a liquid-cooled string PCS energy storage converter. Background Art

[0002] With the widespread application of energy storage systems in grid frequency regulation, peak shaving and valley filling, liquid-cooled string PCS energy storage converters have gradually become the core equipment of large-scale energy storage systems due to their modular design and efficient heat dissipation characteristics. The existing temperature management method of liquid-cooled string PCS energy storage converters usually adopts a unified temperature threshold and a fixed cooling strategy to manage all strings, ignoring the temperature differences and working state differences between different strings. Especially in high temperature environments or high-load operating conditions, the unified cooling strategy cannot meet the personalized cooling needs of different strings, resulting in excessive temperatures in some strings and waste of cooling resources in other strings. In addition, the existing technology rarely considers the coordinated control of cooling resources and power adjustment, and only relies on increasing cooling resources to deal with temperature anomalies, while ignoring the possibility of achieving temperature balance through dynamic power adjustment, making the overall temperature management efficiency of the system low and unable to give full play to the performance advantages of liquid-cooled string PCS energy storage converters. Summary of the invention

[0003] The main purpose of the present invention is to solve the technical problems caused by unreasonable cooling resource allocation and low temperature management efficiency caused by the use of a unified temperature threshold and a fixed cooling strategy in the existing liquid-cooled string PCS energy storage converter temperature management method; A first aspect of the present invention provides a temperature management method for a liquid-cooled string-type PCS energy storage converter, the temperature management method for a liquid-cooled string-type PCS energy storage converter comprising: Collecting temperature data of each string in the liquid-cooled string-type PCS energy storage converter, and calculating temperature characteristic parameters based on the temperature data to obtain a string temperature state matrix; Dividing the cooling priority of each string according to the string temperature state matrix, and allocating cooling resources according to the cooling priority to obtain a dynamic cooling execution instruction set; Based on the dynamic cooling execution instruction set and the string temperature state matrix, a string temperature distribution diagram including hot spot strings and cold spot strings is identified, and a power adjustment scheme is calculated according to the string temperature distribution diagram to obtain a power allocation execution matrix; A corresponding temperature control strategy is selected according to the environmental state parameters of the liquid-cooled string-type PCS energy storage inverter and the power allocation execution matrix, wherein the temperature control strategy is used to set parameters of various components of the cooling system corresponding to the liquid-cooled string-type PCS energy storage inverter to achieve temperature management of the liquid-cooled string-type PCS energy storage inverter.

[0004] Optionally, in a first implementation of the first aspect of the present invention, the temperature data of each string in the liquid-cooled string PCS energy storage converter is collected, and a temperature characteristic parameter is calculated based on the temperature data to obtain a string temperature state matrix, including: The temperature data is collected by using a temperature sensor array arranged at key points of each PCS module and battery cluster of the liquid-cooled string PCS energy storage converter to obtain an original temperature data set; The temperature original data set is differentiated based on the current power state and position information of each string to obtain a string temperature threshold setting matrix; Calculate the difference between the current temperature of each string and the corresponding threshold in the string temperature threshold setting matrix to obtain the temperature safety margin of each string; The temperature difference between adjacent strings and the ratio of the physical distance are calculated to obtain the temperature gradient vector between the strings; The temperature safety margin and the temperature gradient vector are organized in a matrix according to the string numbers to obtain a string temperature state matrix.

[0005] Optionally, in a second implementation of the first aspect of the present invention, the cooling priority of each string is divided according to the string temperature state matrix, and cooling resources are allocated according to the cooling priority, and the dynamic cooling execution instruction set is obtained, including: According to the temperature safety margin in the string temperature state matrix, each string is divided into three levels of cooling priority to obtain a string cooling priority classification table; Based on the string cooling priority classification table, the speed of the main circulation pump of the cooling system of the liquid-cooled string PCS energy storage converter is adjusted and calculated to obtain the main circulation flow control parameter; Calculate the flow distribution ratio of the coolant pipeline section corresponding to each string according to the main circulation flow control parameter and the string cooling priority classification table to obtain a group pipeline control instruction; Calculate the microchannel flow adjustment scheme inside each PCS module based on the grouped pipeline control instructions and the internal heat distribution data of each string to obtain a microchannel control instruction; The main circulation flow control parameters, the grouped pipeline control instructions and the microchannel control instructions are collaboratively optimized to obtain a dynamic cooling execution instruction set.

[0006] Optionally, in a third implementation of the first aspect of the present invention, the main circulation pump speed of the cooling system of the liquid-cooled string PCS energy storage converter is adjusted and calculated based on the string cooling priority classification table to obtain the main circulation flow control parameters including: Performing statistical analysis on the number and temperature safety margin of each priority string in the string cooling priority classification table to obtain the overall cooling demand value of the system; According to the overall cooling demand value of the system, a preset pump speed-flow mapping table is queried to obtain a basic main circulation pump speed value; When the proportion of high priority strings exceeds a threshold, the basic main circulation pump speed value is corrected for emergency cooling demand to obtain a corrected pump speed value; The pump power control curve is calculated based on the corrected pump speed value combined with the system pipeline pressure data to obtain a pump speed control sequence, and the pump speed control sequence is converted into a pulse width modulation signal parameter recognized by the pump controller to obtain the main circulation flow control parameter.

[0007] Optionally, in a fourth implementation of the first aspect of the present invention, the identifying a string temperature distribution diagram including hot spot strings and cold spot strings based on the dynamic cooling execution instruction set and the string temperature state matrix, and calculating a power adjustment scheme according to the string temperature distribution diagram to obtain a power allocation execution matrix includes: Calculate the temperature deviation value of each string according to the string temperature state matrix and the cooling resource allocation in the dynamic cooling execution instruction set to obtain a string temperature deviation table; Based on the string temperature deviation table, the strings with a temperature higher than the average value are marked as hot-spot strings, and the strings with a temperature lower than the average value are marked as cold-spot strings, so as to obtain a string temperature distribution diagram; According to the cooling resource allocation, the amount of power that can be reduced by the hot spot group string and the amount of power that can be increased by the cold spot group string are calculated to obtain the maximum allowable power transfer amount; A power gradient transfer scheme is determined according to the maximum allowable power transfer amount, and hysteresis control parameters are set for the power transfer timing table in the gradient transfer scheme to obtain a power allocation execution matrix.

[0008] Optionally, in a fifth implementation of the first aspect of the present invention, calculating the amount of power that can be reduced by the hot spot group string and the amount of power that can be increased by the cold spot group string according to the cooling resource allocation situation to obtain the maximum allowable power transfer amount includes: Acquire cooling resource allocation data and actual cooling capacity data of each string in the dynamic cooling execution instruction set to obtain a string cooling efficiency table; Calculate the maximum reducible power of each hotspot string according to the string cooling efficiency table and the temperature deviation value of each hotspot string, and obtain the hotspot power adjustable table; Based on the temperature margin and power carrying capacity of each cold spot string, the maximum power that can be increased for each cold spot string is calculated to obtain an adjustable cold spot power scale; Performing matching analysis on the hot spot power adjustable scale and the cold spot power adjustable scale to obtain an initial power transfer matching matrix; A constraint condition check is performed according to the power transfer direction and value in the initial power transfer matching matrix to obtain a maximum allowable power transfer amount.

[0009] Optionally, in a sixth implementation of the first aspect of the present invention, the selecting a corresponding temperature control strategy according to the environmental state parameters of the liquid-cooled string PCS energy storage converter and the power allocation execution matrix includes: Collecting ambient temperature, humidity and prediction data to obtain an ambient state parameter set, and combining the ambient state parameter set with the power allocation execution matrix to perform operation mode identification to obtain the current system operating condition type; According to the current system operating condition type, a corresponding basic strategy group is matched from a preset temperature control strategy library to obtain a candidate temperature control strategy set; Calculating and evaluating the cooling efficiency ratio of each strategy in the candidate temperature control strategy set to obtain a strategy scoring result; A corresponding temperature control strategy is selected based on the strategy scoring result, and control parameters of various components of the cooling system are generated according to the temperature control strategy.

[0010] A second aspect of the present invention provides a temperature management system for a liquid-cooled string-type PCS energy storage converter, the temperature management system for the liquid-cooled string-type PCS energy storage converter comprising: A temperature monitoring module is used to collect temperature data of each string in the liquid-cooled string-type PCS energy storage converter, and calculate temperature characteristic parameters based on the temperature data to obtain a string temperature state matrix; A resource allocation module, used to divide the cooling priority of each string according to the string temperature state matrix, and allocate cooling resources according to the cooling priority to obtain a dynamic cooling execution instruction set; A power balancing module, configured to identify a string temperature distribution diagram including hot-spot strings and cold-spot strings based on the dynamic cooling execution instruction set and the string temperature state matrix, and calculate a power adjustment scheme according to the string temperature distribution diagram to obtain a power allocation execution matrix; A strategy execution module is used to select a corresponding temperature control strategy according to the environmental state parameters of the liquid-cooled string-type PCS energy storage inverter and the power allocation execution matrix, wherein the temperature control strategy is used to set parameters of various components of the cooling system corresponding to the liquid-cooled string-type PCS energy storage inverter to achieve temperature management of the liquid-cooled string-type PCS energy storage inverter.

[0011] The temperature management method and system of the above-mentioned liquid-cooled string-type PCS energy storage converter collects temperature data of each string in the liquid-cooled string-type PCS energy storage converter and calculates temperature characteristic parameters to obtain a string temperature state matrix; divides the cooling priority of each string according to the string temperature state matrix and performs cooling resource allocation to obtain a dynamic cooling execution instruction set; identifies hot spot strings and cold spot strings based on the dynamic cooling execution instruction set and the string temperature state matrix and calculates the power adjustment plan to obtain a power allocation execution matrix; selects the corresponding temperature control strategy according to the environmental state parameters and the power allocation execution matrix, and sets parameters for each component of the cooling system. The present invention combines dynamic cooling resource allocation, power dynamic balance and environmental adaptability temperature control strategy to solve the problems of unreasonable cooling resource allocation and low temperature management efficiency, and realizes precise temperature management of liquid-cooled string-type PCS energy storage converters under various working conditions.

[0012] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0013] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a schematic diagram of a first embodiment of a temperature management method for a liquid-cooled string PCS energy storage converter according to an embodiment of the present invention; Figure 2 Schematic diagram of an embodiment of a temperature management system for a liquid-cooled string PCS energy storage converter in an embodiment of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0017] To facilitate understanding of this embodiment, a temperature management method for a liquid-cooled string PCS energy storage converter disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, the method comprises the following steps: 101. Collect temperature data of each string in the liquid-cooled string-type PCS energy storage converter, and calculate temperature characteristic parameters based on the temperature data to obtain a string temperature state matrix; In one embodiment of the present invention, the temperature data of each string in the liquid-cooled string PCS energy storage converter is collected, and the temperature characteristic parameters are calculated based on the temperature data to obtain a string temperature state matrix, including: collecting temperature data through a temperature sensor array arranged at key points of each string PCS module and battery cluster of the liquid-cooled string PCS energy storage converter to obtain a temperature original data set; performing differential processing on the temperature original data set based on the current power state and position information of each string to obtain a string temperature threshold setting matrix; performing difference calculation between the current temperature of each string and the corresponding threshold in the string temperature threshold setting matrix to obtain a temperature safety margin of each string; calculating the temperature difference of adjacent strings and the ratio of the physical distance to obtain a temperature gradient vector between strings; and organizing the temperature safety margin and the temperature gradient vector in a matrix according to the string number to obtain a string temperature state matrix.

[0018] Specifically, first, the temperature data is collected through the temperature sensor array arranged at the key points of each string PCS module and battery cluster of the liquid-cooled string PCS energy storage converter to obtain the original temperature data set. Each string of the converter contains several key heat sources, including IGBT modules, inductors, busbars, and matching battery clusters, so the layout of sensors needs to cover these areas that are prone to heat generation or heat-affected. Sensor types include thermocouples installed on the surface of the IGBT heat sink, optical fiber temperature sensors fixed to the battery cluster housing or terminals, temperature sensors attached to the busbar joints, and protective thermistors that can monitor the external ambient temperature. Each sensor uploads temperature data at fixed time intervals through a distributed sampling node or bus network, and the acquisition frequency is set according to the thermal inertia of the component, so as to ensure sufficient resolution of thermal changes without excessive data accumulation. Each sensor is calibrated before installation to ensure that reliable data can still be output in high current, electromagnetic interference and humidity fluctuation environments. The raw data collected by the sensor is sent to the central measurement and control unit of the converter. The unit marks the corresponding sensor ID and timestamp for each temperature data, and stores these records in the temperature raw data set for subsequent query, comparison or further calculation. The raw data set retains the sampled values ​​that have not been denoised or corrected, and has a complete record of abnormal fluctuations, so that the specific temperature readings and sensor positions can be traced back whenever necessary. In order to cover a variety of working scenarios, the sensor array also considers the spatial distribution at all levels inside the container to ensure that the hot spot temperature in the high-load area can be obtained and the overall heat diffusion in the low-power area can be monitored. Information transmission uses a communication protocol with error correction and verification functions. If the sensor communication is disordered or the data conflicts, the measurement and control unit will record the abnormal identification in the temperature raw data set. The temperature raw data set formed in this way can be output to the subsequent processing program in a timed or triggered manner. The detection position of each sensor is also associated with the actual physical coordinates through a set of mapping tables for further differential analysis or model calculation. The whole process ensures the comprehensiveness of the temperature data and the integrity of the acquisition process.

[0019] Next, the temperature original data set is differentiated based on the current power state and location information of each string to obtain the string temperature threshold setting matrix. The power state here indicates the charge and discharge amplitude, load duration and corresponding current fluctuation of each string at a certain moment, and the location information indicates the location characteristics of the environment in which the string is located, such as close to the heat dissipation duct, away from the liquid cooling pipe or close to the container wall. When performing differentiated processing, the system will first read the power history of each string, including the average power size and instantaneous peak current in the previous period, and combine the coordinate label of the string in the physical space to determine whether it is in a section prone to heat concentration. If a string is in a high power output or full load state for a long time in the near future, the upper limit of the reference temperature corresponding to its sensor is tightened in the threshold setting link, so that the string has a more stringent temperature critical value, so as to timely reflect the greater heat dissipation pressure that the string may face. On the contrary, for strings with stable power state and good heat dissipation position, the system reserves a relatively loose temperature allowable range for them. During this process, each sensor is assigned a weighting coefficient, which consists of a power weight and a position weight. These weights not only refer to the historical power state, but also take into account whether there is dense internal component stacking or high thermal interference from surrounding strings. Once the weighting is completed, the system will correct the baseline temperature threshold of each sensor to make the threshold of nodes with greater thermal risks lower to improve monitoring sensitivity. All corrected thresholds are arranged into a multidimensional matrix, with each row or column identifying a specific sensor or string, and the matrix elements are the corresponding temperature thresholds. For ease of indexing and updating, the matrix also contains a time dimension label to ensure that the threshold recalculation can be triggered every time the power state is updated. The string temperature threshold setting matrix formed in this way can be flexibly adjusted under different working conditions, and will not cause high-load string monitoring lag or low-load string monitoring to be too strict due to a single fixed threshold. This differentiated processing provides the basis for subsequent data comparison and safety margin assessment.

[0020] Subsequently, the difference between the current temperature of each string and the corresponding threshold in the string temperature threshold setting matrix is ​​calculated to obtain the temperature safety margin of each string. This step extracts the current average temperature or key sensor temperature of each string from the original temperature data set, and then matches it to the corresponding threshold entry in the threshold setting matrix generated in the previous step. The difference between the two is obtained by a simple subtraction operation. The difference is the temperature safety margin. If the safety margin remains at a high level, it means that the corresponding string still has a significant heat dissipation margin and will not trigger an overtemperature alarm. If the safety margin approaches zero or a negative value appears, it means that the current temperature of the string is close to or even exceeds the dynamic threshold, and further cooling or power limiting measures need to be taken. When implementing the calculation, the system will calculate the difference for all temperature sensors under each string one by one, and summarize them into a safety margin set in the internal data structure to accurately reflect the thermal health of each component inside the string. Some data anomalies, such as a sudden and large jump in the temperature reading of a sensor, are combined with the previous change trend of the sensor to determine whether it truly reflects the thermal shock. If there is indeed a high-speed temperature rise phenomenon, the safety margin will be lowered in real time. If it is determined to be a transient interference, the system will mark this reading as a suspicious value, but it will still be retained in the calculation trajectory of the safety margin to avoid discarding potential information that may indicate a component failure. The safety margin is not only a simple set of values, but can also be expanded into a temperature trend curve. By observing the decrease or gain of the margin over time, it is helpful to locate the parts where heat dissipation is not smooth. The whole process is designed to convert the raw temperature data into an intuitive indicator of danger, so that temperature deviations can be discovered and processed more quickly, while also taking into account stability and accuracy. After the difference calculation is completed, the system generates a safety margin list with a number and a timestamp for each string. The margin value of each string can be directly retrieved when querying, thereby helping operation and maintenance personnel or automatic control logic to grasp the thermal response of the string in a more timely manner.

[0021] Next, the temperature difference of adjacent strings and the ratio of the physical distance are calculated to obtain the temperature gradient vector between strings. This step first obtains the adjacent relationship and the corresponding spatial distance of each string through topological mapping. The spatial distance can be the straight-line distance between the center points of the strings, or the effective heat dissipation path formed by factors such as partitions and pipelines in the actual structure. Subsequently, the system obtains the temperature values ​​of adjacent strings at the same time from the original temperature data or the calculated average temperature, and subtracts the two to obtain the temperature difference. Using the aforementioned physical distance, the temperature difference is divided by the distance to obtain a numerical gradient, and the positive or negative gradient indicates which side has a higher temperature. After the calculation is completed, the gradient is combined into a vector form and recorded, for example, marked as "temperature gradient vector from string A to string B", which is used to describe possible heat migration trends. If the distance between the two strings is close and the temperature difference is large, the gradient vector will reflect a higher heat flux density, which is particularly obvious when multiple strings are loaded at the same time. In order to update dynamically, this calculation is kept at the same frequency as the temperature acquisition, so that if a gradient is observed to rise significantly at a certain moment, it can be said that one side has begun to heat up rapidly or the other side is not cooling enough. In some container designs, the direction of the liquid cooling channel will also affect the lateral transfer of heat. If two adjacent strings share the same liquid cooling branch, the system will specifically mark whether there is a common cooling circuit when recording the gradient vector for reference when subsequently allocating cooling resources. The gradient vector also helps to identify the actual heat source center. For example, if a string shows a high positive gradient in multiple directions, it means that the temperature of the string is much higher than that of the adjacent string, and there will be a more obvious trend of heat diffusion. This numerical gradient measurement reduces the error of relying on manual experience to judge the risk of heat diffusion. A series of adjacent string temperature differences and distance ratios obtained after calculation will be stored in a data table, which can quickly index the thermal gradient strength of any pair of strings when querying. In this way, a real-time heat transfer map can be formed in the system, providing a more targeted reference for overall temperature management.

[0022] Finally, the temperature safety margin and the temperature gradient vector are organized in a matrix according to the string number to obtain the string temperature state matrix. Starting from the safety margin list of each string and the gradient vector data table corresponding to the adjacent string, the process assigns a unique row or column identifier to each string, records the safety margin value in the corresponding unit, and then fills in the relevant temperature gradient information in the interactive unit with the adjacent string. If a string has a gradient with multiple adjacent strings, multiple non-zero fields will appear in the matrix structure, indicating different degrees of heat exchange trends between each other. The matrix organization method can simultaneously display the absolute temperature risk (i.e., safety margin) and the relative temperature relationship (i.e., gradient vector) of each string in a table, so that any analysis or calculation module can retrieve all thermal characteristic values ​​in one place. To ensure the timeliness of the data, the system adds the timestamp correction of each string and the adjacent relationship when writing the matrix, and uniformly calibrates the exact time of data update through the internal clock to avoid inconsistent matrix data due to differences in acquisition timing. In a large-scale scenario with multiple strings in parallel, this matrix can reflect which strings are in a critical state close to the threshold and whether they will have a significant thermal load impact on adjacent strings. The safety margin and gradient information are stored in different hierarchical indexes in the matrix for differentiated reading; the safety margin item is used to judge the degree of proximity of the temperature to the respective thresholds, and the gradient item is used to judge the intensity of heat propagation between adjacent strings. Such a data structure can be directly read and incorporated into various management and control strategies after real-time updates, combining absolute temperature with relative heat diffusion information to form a complete picture covering static thermal risks and dynamic thermal interactions. At this point, the string temperature state matrix is ​​successfully generated in this embodiment. By integrating the safety margin of each string and the gradient vector between adjacent strings, it provides an intuitive and efficient method for the system to understand and characterize the thermal distribution conditions within the liquid-cooled string PCS energy storage inverter.

[0023] 102. Divide the cooling priority of each string according to the string temperature state matrix, and allocate cooling resources according to the cooling priority to obtain a dynamic cooling execution instruction set; In one embodiment of the present invention, the cooling priority of each string is divided according to the string temperature state matrix, and cooling resources are allocated according to the cooling priority to obtain a dynamic cooling execution instruction set, including: dividing each string into three levels of cooling priority according to the temperature safety margin in the string temperature state matrix to obtain a string cooling priority grading table; adjusting and calculating the main circulation pump speed of the cooling system of the liquid-cooled string PCS energy storage inverter based on the string cooling priority grading table to obtain a main circulation flow control parameter; calculating the flow distribution ratio of the coolant pipeline section corresponding to each string according to the main circulation flow control parameter and the string cooling priority grading table to obtain a group pipeline control instruction; calculating the microchannel flow adjustment scheme inside each PCS module based on the group pipeline control instruction and the internal heat distribution data of each string to obtain a microchannel control instruction; and collaboratively optimizing the main circulation flow control parameter, the group pipeline control instruction and the microchannel control instruction to obtain a dynamic cooling execution instruction set.

[0024] Specifically, the three-level cooling priority is divided for each string according to the temperature safety margin in the string temperature state matrix. The process of obtaining the string cooling priority classification table starts with reading the temperature safety margin value of each string. These values ​​contain a measure of the difference between the actual temperature and the acceptable threshold, thereby reflecting the pressure faced by each string at the heat dissipation level. At this time, the system will scan all strings and divide each string into three levels: emergency cooling group, high-demand cooling group, and conventional cooling group according to the pre-designed temperature margin partitioning method. The temperature safety margin of the emergency cooling group is lower than a certain critical value, indicating that its temperature is very close to the set threshold; the margin of the high-demand cooling group is slightly higher than the emergency interval, but still shows a certain degree of temperature rise pressure; the conventional cooling group has a relatively ample temperature margin. In order to form a clear and executable classification table, the system will list the IDs and surplus values ​​of all strings after scanning, and mark the corresponding priority labels next to them. Each priority section has a clearly defined value range. For example, if the surplus is less than a certain limit value, it belongs to the emergency cooling group, and if the surplus is large enough, it is classified into the conventional cooling group. The middle part is automatically assigned to the high-demand cooling group. Such a classification process is usually executed within a fixed time period or triggered immediately when a sudden drop in the temperature surplus of a string is detected, so as not to miss the timely identification of emergency conditions. The system records the classification results of each string with a timestamp and stores them in a callable database structure to retrieve the priority allocation information at any time. If the same string has been in the emergency cooling level for a long time, it means that both cooling measures and temperature warnings need to be strengthened. Therefore, this classification table can assist all subsequent temperature management operations to make targeted decisions. The three-level cooling priority classification table is essentially a mapping list, with each row listing the string number, real-time temperature safety margin, classification category and update time. Once the grading table is generated, it can be compared with other data sources, such as the temperature trend information of other measuring points near the string, to determine whether there is regional warming or local heat dissipation difficulties. After the grading table is established, there is no need to repeatedly present the absolute temperature or heat dissipation details of each string. Only the priority label can be retained in the same table to complete the preliminary preparations for a large number of cooling decisions. The grading table directly reflects the thermal pressure relationship between each string and its environment, and can find the objects that need cooling most through quick indexing. The thresholds and intervals of the grading are often configured based on the specific conditions of the site and the heat resistance level of the equipment. Fixed intervals can be used or dynamic fine-tuning can be made based on field experience. Once the table is updated, it will immediately enter the next stage of cooling allocation calculations.

[0025] Based on the string cooling priority classification table, the speed of the cooling system main circulation pump of the liquid-cooled string PCS energy storage converter is adjusted and calculated. In the process of obtaining the main circulation flow control parameters, the system will first collect the maximum flow and corresponding pump power range that the current cooling system can provide. Under the priority reference provided by the classification table, the system can determine how many strings are in the emergency cooling group or the high-demand cooling group, so as to preliminarily estimate the overall cooling demand. If more strings fall into the emergency cooling group, it means that the entire system requires a higher intensity of heat dissipation capacity. At this time, the main circulation pump speed often needs to be increased to a relatively high level so that the coolant can circulate quickly in the pipeline and take away more heat. If most of the strings are in the conventional cooling group, the requirements for the pump speed are lower, which can meet the heat dissipation and reduce the energy consumption of the cooling system itself. In order to perform more accurate calculations, the system often maintains a pump speed-flow mapping table, which lists the theoretical flow that the main circuit can achieve under different pump speed settings and the average flow range during actual measurement, and is accompanied by data such as pressure loss. The algorithm will add the number of emergency cooling groups and high-demand cooling groups and combine the depth of the temperature safety margin of each string to calculate an overall cooling demand index, and then compare it with the pump speed-flow mapping table to directly find the basic pump speed value that can match the demand. After the priority distribution is statistically calculated, once it is found that the number of emergency cooling groups exceeds the set threshold, the system will introduce a correction coefficient to increase the basic pump speed value upward to meet the centralized emergency cooling requirements. After that, the system will further fine-tune the pump speed in combination with the pipeline pressure condition returned by the loop pressure monitor. The main circulation flow control parameters obtained in this way usually include a target speed or speed range and the corresponding power limit, which are used to instruct the variable frequency drive module of the main pump. When the circulating pump reaches the new speed, the system will also observe the pressure changes of the liquid cooling circuit and whether the temperature of each string has a rapid decline trend in a short time. If there are still many strings at the emergency cooling level, the system will increase the main circulation flow again until the number of emergency cooling groups is significantly alleviated or the pump speed reaches the maximum limit that can be tolerated. This adjustment process will be executed multiple times within a certain time window, and by constantly comparing the latest situation of the classification table, it will decide whether to continue to increase or slightly reduce the pump speed. The main circulation flow control parameters obtained in this way can not only adapt to the sudden change of local heat dissipation urgency, but also save energy when the cooling demand is reduced, thereby achieving more effective scheduling of liquid cooling resources.

[0026] According to the main circulation flow control parameters and the string cooling priority classification table, the flow distribution ratio of the coolant pipeline section corresponding to each string is calculated. The process of obtaining the group pipeline control instruction will regard the total flow output of the main circulation pump as an allocable resource, and then allocate each branch in combination with the priority information obtained from the classification table. In order to clarify the cooling channel of each string, the system often connects the inlet and outlet of several strings in parallel to the same branch pipeline in the design, and the branch pipeline is connected to the main circulation pump through a regulating valve or a proportional valve. When the number of emergency cooling groups is large, it is necessary to allocate a higher flow on the branch where these strings are located. The common way is to increase the opening of the corresponding branch valve or reduce the diversion ratio of other non-emergency branches. In the specific implementation, the system first retrieves the string ID of the emergency cooling group from the classification table, and marks the corresponding branch pipeline as a "high priority branch", marks the pipeline where the high-demand cooling group is located as a "medium priority branch", and the pipeline where the conventional cooling group is located is regarded as a "low priority branch". The system then obtains the total flow currently available in the main circulation flow control parameters, determines its minimum proportion and upper limit proportion according to the number of high-priority branches, and then calculates a target flow or opening percentage for each pipeline based on the demand conditions of the medium-priority and low-priority branches. In order to prevent the phenomenon that a high-priority branch monopolizes the flow and causes other branches to be seriously insufficiently cooled, the system usually sets a restriction condition to ensure that each pipeline in the main circulation can maintain a certain minimum flow to maintain the basic heat dissipation cycle. After the allocation is completed, the system outputs a set of grouped pipeline control instructions, which are often the target opening or pulse modulation mode data for each branch valve. The valve controller will adjust the valve position according to the instructions and send the information back to the main control module in real-time operation. When it is found that the temperature of a group string in a high-priority branch drops to a certain range, the system will update the classification table and iterate the allocation algorithm again to ensure that the flow distribution of each branch always fits the current priority layout. Different circuits may also refer to the pressure or flow rate sensors installed on site when allocating flow. If the pressure of a pipeline fluctuates greatly, the corresponding compensation adjustment will be triggered. In this way, each pipeline can obtain the liquid cooling supply that matches its current cooling needs, and the allocation instructions are coordinated in timing with the speed control of the main circulation pump to form a multi-level dynamic cooling management architecture.

[0027] Based on the grouped pipeline control instructions and the internal heat distribution data of each string, the microchannel flow adjustment scheme inside each PCS module is calculated. In the process of obtaining the microchannel control instructions, the system's focus is further explored from the pipeline level to the inside of the module. This type of converter usually designs a microchannel cold plate under components such as IGBT modules, heat dissipation substrates or inductors, in order to achieve more efficient cooling around high heat density components. When the main circulation and branch pipelines are allocated with flow, the amount of coolant entering the module is generally guaranteed, but the flow distribution between each microchannel still needs to be further fine-tuned so that the area with the most concentrated heat source can get more liquid cooling contact first. This microchannel flow adjustment is often completed by micro valves or adjustable throttling mechanisms, and the control methods include analog signal adjustment or digital proportional valve control. In order to perform calculations, the system must first understand the temperature conditions of several core heat dissipation surfaces inside the module from the heat distribution data. These data are usually provided by small temperature sensors arranged near IGBTs or inductors. By comparing the values ​​of these sensors, the system can determine which area has the highest heat dissipation pressure. The system then fits the local increase and decrease strategy of the microchannel network with reference to the flow upper limit in the grouped pipeline control instruction. For example, if the temperature of the IGBT area rises fastest, the microchannel control instruction tends to increase the opening of the flow channel valve in this area, while reducing the flow of other relatively temperature-safe parts. In order to prevent excessive flow in a single channel from causing insufficient flow in the downstream area, the control algorithm often constrains and optimizes the channels inside the entire module, and achieves differentiated cooling of different heat sources through local distribution while keeping the overall flow unchanged. This process can cope with regional high heat caused by high transient loads or large current shocks, and can also reduce the consumption of excessive cooling when the heat dissipation demand decreases. When the microchannel control instruction is issued, the small valves inside the module open and close according to the adjustment ratio of the instruction, and the coolant is distributed to the parts that need cooling most urgently in real time. Microchannel adjustments are often repeated in very short cycles to avoid heat accumulation in local hot spots. The system then collects the sensor temperature value of the module again to evaluate the actual cooling effect, and will correct the microchannel allocation parameters in the next round of scheduling. This ensures that the flow distribution within each PCS module truly reflects the current heat load characteristics and achieves more refined heat dissipation management.

[0028] The main circulation flow control parameters, the grouped pipeline control instructions and the microchannel control instructions are collaboratively optimized to obtain the dynamic cooling execution instruction set. The process is uniformly managed at the system level. In order to avoid conflicts or over-responses between single instructions, the algorithm usually integrates and adapts these instructions. First, the system reads the target pump speed and the upper limit of the overall flow set in the main circulation flow control parameters, and then analyzes the opening ratio of the grouped pipeline control instructions on this basis to check whether there is a phenomenon of over-allocation or insufficient flow of some pipelines due to priority conflicts, and makes a relative balance for the opening of some valves when necessary. Then the system will enter the inspection program at the microchannel level, and judge whether to make additional adjustments to the flow of the grouped pipeline according to the local heat load reflected by the microchannel control instructions, so as to avoid the situation where the local microchannel has already increased the flow and still maintains the maximum opening of the high priority branch, thereby causing an imbalance in the overall flow distribution or high-load operation of the circulating pump. In the specific implementation of collaborative optimization, the algorithm usually follows the principle of macro first and micro later, first using the pump speed and grouped valves to determine the distribution of the overall heat dissipation capacity of the system, and then using the microchannel strategy to make fine corrections to the terminal. If it is found that the high-priority branch has significantly reduced the local temperature through microchannel measures, while other branches continue to be undercooled, the system will make corresponding flow callbacks for the group pipeline control instructions, and re-allocate resources for the branches where the temperature is still high, so that the string temperature in the entire converter can converge to the safe range. The mutual influence between the instructions at each level is realized in the integrated optimization algorithm. The algorithm mainly selects the optimal overall matching solution based on whether there are still a large number of strings showing low margin or rapid temperature rise tendency in the temperature state matrix, and combines the real-time power consumption and pressure limit of the cooling system. The final dynamic cooling execution instruction set will be written into a structured instruction document or data packet, which contains specific information such as pump speed set value, group pipeline valve opening, microchannel adjustment ratio, and lists the order of effectiveness within the specified time period. Once the instruction set is issued, the controllers at all levels execute the operation according to the established protocol, and continuously send back new temperature and pressure data during the execution process, so as to perform collaborative optimization again in the next cycle.

[0029] Furthermore, the main circulation pump speed of the cooling system of the liquid-cooled string PCS energy storage inverter is adjusted and calculated based on the string cooling priority grading table to obtain the main circulation flow control parameters, including: statistical analysis of the number of strings of each priority level in the string cooling priority grading table and the temperature safety margin to obtain the overall cooling demand value of the system; querying a preset pump speed-flow mapping table according to the overall cooling demand value of the system to obtain a basic main circulation pump speed value; when the proportion of high-priority strings exceeds a threshold, the basic main circulation pump speed value is corrected for emergency cooling demand to obtain a corrected pump speed value; based on the corrected pump speed value and combined with the system pipeline pressure data, a pump power control curve is calculated to obtain a pump speed control sequence, and the pump speed control sequence is converted into a pulse width modulation signal parameter recognized by a pump controller to obtain a main circulation flow control parameter.

[0030] Specifically, the technical solution first statistically analyzes the number and temperature safety margin of each priority string in the string cooling priority classification table to obtain the overall cooling demand value of the system. The implementation process starts with reading the existing classification table, which usually contains several record rows, each row identifying the real-time priority level of a string and the corresponding temperature safety margin. The system traverses all records in a scanning manner. When encountering a string marked as a high priority, its temperature safety margin is multiplied by the internally defined severity coefficient and then accumulated. Similarly, the temperature safety margin of the medium priority string will also be multiplied by the weight constant of another level and added to the sum, while the low priority string usually does not significantly increase the global cooling demand, so only sporadic numerical accumulation is performed or its impact is directly ignored. In this way, the pressure level of resources faced by each priority segment at the temperature level can be obtained. When counting, the number of high priority strings and medium priority strings will be paid attention to. If the number of high priority strings is particularly large, a higher weighting factor will be given in the final synthesis to reflect its urgency for cooling. In order to more accurately characterize the difference in heat dissipation load between different strings, the system will also reference a time decay function for each priority string, and assign a moderately weighted reduction coefficient to the strings that remain in a high temperature state for a long time but are not supercritical, so that the truly urgent strings can occupy the main weight in the comprehensive statistics. Statistical algorithms sometimes use piecewise functions or linear-exponential mixed functions to complete the mapping of temperature safety margins. For example, a nonlinear accumulation process can be set so that extremely low safety margins bring higher increments when accumulated. For example, if the safety margin of a high-priority string is The safety margin for medium priority is , then the following relationship can be set in the statistical process: HP represents a high priority set, and MP represents a medium priority set. , ,γ, is a constant used to adjust the weight and sensitivity. By calculating this formula, the comprehensive cooling demand of strings of different priorities at the current moment can be quantified. Finally, the above sum is defined as the overall cooling demand value of the system, forming a scalar that can be directly called by subsequent modules. When it is detected that there are signs of rapid temperature rise in high-priority strings, the statistical algorithm will continue to monitor the evolution trend of its safety margin and dynamically increase or decrease the relevant weight during the periodic refresh process. If some strings still need to be cooled urgently after several refreshes, the strings will be individually marked, and the possible heat dissipation bottlenecks will be found through a small range search and the information will be provided to the next stage of the flow allocation mechanism. This overall cooling demand value can be stored synchronously in an easily accessible data area, with time stamps and priority tags, to trace back the current system at any time. The heat dissipation pressure provides a basis for liquid cooling scheduling. In this way, the integrated statistics of the number of strings and temperature safety margins in the priority classification table are completed, and the overall cooling demand value of the system is successfully output.

[0031] Next, the preset pump speed-flow mapping table is queried according to the overall cooling demand value of the system to obtain the basic main circulation pump speed value. The implementation process will first load a mapping table compiled during the experiment or field test phase. The table is often presented in a two-dimensional or three-dimensional structure: the horizontal axis corresponds to the pump speed, the vertical axis corresponds to the net flow of the system, and the third dimension may record the corresponding rated power consumption or pipeline pressure. The system uses the overall cooling demand value obtained in the previous step to find a matching interval in the table. The process usually includes an interpolation calculation to locate a more accurate speed setting value between the discretized pump speed marks. The construction of the mapping table relies on a large amount of test data, covering the actual flow performance under low-speed, high-speed and medium-speed conditions, and combining different pipeline layouts and ambient temperature factors. The system compares the cooling demand value with the demand threshold listed in the table, and reads the corresponding pump speed entry when the mapping row closest to the demand value is found. For example, when the demand value is low, a relatively low speed will be matched to maintain the necessary liquid cooling cycle and avoid unnecessary energy waste; when the demand value is close to the highest interval, a higher speed is read from the table, corresponding to the large flow mode. In order to make the result more refined, the interpolation algorithm often needs to use piecewise polynomials or bicubic splines, allowing smoother speed values ​​to be calculated between known discrete measurement points. For example, when the required value is between m and n, the following interpolation function with weighted coefficients can be used for calculation: in and are the rotation speeds of adjacent measuring points, and are the corresponding upper and lower boundaries of the demand value interval, is a smoothing factor used to fine-tune the shape of the interpolation curve. This is the basic main circulation pump speed value. When querying the mapping table and performing interpolation, the system ensures that the pipeline structure is consistent with the structure targeted by the mapping table, and checks the deviation correction coefficient of the ambient temperature or coolant temperature before reading the data. If the temperature difference exceeds a certain range, the coefficient will be incorporated into the interpolation algorithm to make the final speed value closer to the real-time working condition. The output of this stage is a specific value or a set of upper and lower limits of the interval, which represents the most suitable basic pump speed at the moment, and is used to support subsequent pump speed correction operations.

[0032] When the proportion of high-priority strings exceeds the threshold, the basic main circulation pump speed value is corrected for the emergency cooling demand to obtain the corrected pump speed value. The implementation process will first scan the overall proportion of high-priority strings from the classification table, and compare this proportion with the pre-defined threshold. If the proportion is still in the normal range, the basic pump speed value does not need to be modified. If the proportion exceeds the threshold, it means that there is a relatively intensive and urgent heat dissipation demand in the current period, and an incremental correction factor needs to be superimposed on the basic pump speed value. The calculation of the correction factor can be implemented by a linear or nonlinear function. If a linear function is selected, the system gives a proportional coefficient and multiplies the proportion of high-priority strings to obtain a correction increment. If a nonlinear method is selected, a speed increase curve can be defined that increases exponentially or stepwise with the proportion. When the proportion is high, the correction amount reaches a large amplitude, so that the pump speed is significantly increased in a short time, thereby providing more flow for the most urgent heat dissipation object. When performing the correction, the mechanical characteristics of the pump, such as the maximum safe speed and acceleration limit, must also be considered. The system will limit the correction amount so that the pump does not jump directly to the limit value and produce unnecessary impact. At the same time, after each correction calculation, the pipeline pressure loss and heat removal speed that may be caused by the current flow will be included in the review process. If the proportion of high-priority strings is too high but the actual temperature has been alleviated, no further correction will be made. In some algorithms, the corrected pump speed will be maintained for a period of time before a second judgment is made. If the temperature of the high-priority string still does not fall, a new round of correction will be performed again. The corrected pump speed value formed in this way is defined to be closer to the real-time heat dissipation situation than the basic pump speed value, which is manifested in the ability to quickly increase the supply capacity of the liquid cooling circuit within a short period of time. Monitoring the proportion of high-priority strings can ensure a rapid response to local hot spots, especially in periods of sudden large power changes or sharp rises in ambient temperature. It plays a role in supplementing cooling protection. Once the corrected pump speed value is confirmed, it will be stored in a special data segment and marked as the actual set value at the current moment to complete the final pump scheduling preparation.

[0033] Based on the corrected pump speed value combined with the system pipeline pressure data, the pump power control curve is calculated to obtain the pump speed control sequence, and the pump speed control sequence is converted into a pulse width modulation signal parameter recognized by the pump controller to obtain the main circulation flow control parameter. The implementation process starts from the corrected pump speed value, first querying the pipeline pressure or pressure distribution data recorded in real time in the system to determine whether the pressure when the pump is running at this speed will exceed the safe working range. In many cases, a set of pressure detection points are used to reflect the resistance load in the pipeline. If the resistance is low, the power demand of the pump will not rise excessively after the speed is increased; on the contrary, if the resistance is large, it may be necessary to make fine adjustments on the power distribution side to prevent the pump from entering an overcurrent or overload state. According to the corrected pump speed value v_ref and the pipeline pressure p_line, the system will construct a power control curve, which often gives a piecewise function for different pressure segments to specify the pump torque or load current. When the pressure rises beyond a certain threshold, the power upper limit will be increased on the curve to maintain the established flow target. In practical operation, the curve consists of a set of point sequences (pressure, power) and corresponding control strategies. The pump speed control sequence can be defined as one or more linear functions, or a step function with overload protection. After the pump speed control sequence is generated, the speed targets under the sequence are converted into pulse width modulation (PWM) parameters through reverse table lookup or built-in driver mapping program. Pump controllers often rely on PWM signals with fixed duty cycle and frequency for speed regulation. Therefore, the system must disassemble the speed control sequence into a series of discrete time points, each of which is accompanied by a PWM duty cycle value and possible acceleration time schedule. This enables the pump to switch smoothly from the original speed to the new target speed and select the matching power curve segment under different pressure conditions. The final output is the main cycle flow control parameters, including the entire time sequence and the corresponding PWM configuration. When the control program starts to execute, the pump immediately follows the sequence to gradually increase or decrease the speed. At the same time, the pressure sensor feedback value is checked after each segment is completed. If the actual pressure or flow deviates too much from the expected, it will be dynamically fine-tuned in the next time slice. This set of control parameters can be directly sent to the variable frequency drive unit, and the data link with the monitoring center can be kept open so that it can be refreshed again when the cold and hot loads suddenly change. This completes the hardware signal mapping from the corrected pump speed to the final execution, so that the cooling system of the liquid-cooled string PCS energy storage converter can obtain reasonable and effective main circulation flow distribution.

[0034] 103. Identify a string temperature distribution diagram including hot-spot strings and cold-spot strings based on the dynamic cooling execution instruction set and the string temperature state matrix, and calculate a power adjustment plan according to the string temperature distribution diagram to obtain a power allocation execution matrix; In one embodiment of the present invention, the identification of a string temperature distribution diagram including hot spot strings and cold spot strings based on the dynamic cooling execution instruction set and the string temperature state matrix, and the calculation of a power adjustment plan based on the string temperature distribution diagram to obtain a power allocation execution matrix includes: calculating the temperature deviation value of each string based on the string temperature state matrix and the cooling resource allocation in the dynamic cooling execution instruction set to obtain a string temperature deviation table; marking the strings with a temperature higher than the average value as hot spot strings and marking the strings with a temperature lower than the average value as cold spot strings based on the string temperature deviation table to obtain a string temperature distribution diagram; calculating the amount of power that can be reduced by the hot spot string and the amount of power that can be increased by the cold spot string based on the cooling resource allocation to obtain a maximum allowable power transfer amount; determining a power gradient transfer plan based on the maximum allowable power transfer amount, and setting hysteresis control parameters for the power transfer timing table in the gradient transfer plan to obtain a power allocation execution matrix.

[0035] Specifically, the temperature deviation value of each string is calculated according to the string temperature state matrix and the cooling resource allocation in the dynamic cooling execution instruction set. The process of obtaining the string temperature deviation table will first obtain the actual temperature or equivalent average temperature of each string at the current moment, and read the cooling resource information allocated to the string from the dynamic cooling execution instruction set. The temperature state matrix records the temperature safety margin, temperature gradient and mutual heat conduction characteristics of each string, and the resource allocation in the dynamic cooling execution instruction set includes data such as flow, valve opening or microchannel adjustment. When calculating the temperature deviation value, the system will make corrections based on the difference between the cooling resources received by each string and its temperature rise or fall rate, so as to avoid imbalanced judgment caused by relying solely on temperature readings. The implementation method can obtain a temperature deviation benchmark by subtracting the theoretical temperature expectation of the string from the actual temperature, and then calculate the correction value in combination with the cooling potential brought by the cooling resource allocation. When the resource allocation is large but the actual temperature is still high, the deviation value will show a significant positive deviation, indicating insufficient heat dissipation or excessive heat source intensity; if the resource is small and the temperature remains at a low level, the deviation value is negatively distributed, suggesting that the heat burden of the string is limited or the surrounding environment has a good cooling effect. After counting these differences, a temperature deviation table is generated. The table lists the temperature deviation values ​​according to the string number and can be accompanied by the cooling resource allocation status of each string. If an extreme positive deviation occurs, it usually indicates that the current cooling strategy cannot meet the thermal requirements of the string. Before outputting the temperature deviation table, the system will also filter or mark abnormal readings, such as sensor failure or data jitter time is too short, to maintain the availability of the record entries in the table. Each deviation value is written to the database together with the timestamp for quick retrieval in the subsequent temperature distribution analysis and power adjustment process.

[0036] Based on the string temperature deviation table, the strings with temperatures higher than the average are marked as hot spots, and the strings with temperatures lower than the average are marked as cold spots. The process of obtaining the string temperature distribution map takes the temperature deviation table as input and extracts the current deviation value of each string from it. The system first calculates a global average temperature deviation by statistically analyzing all deviation values, which is used to divide the baseline of hot spots and cold spots. For each string, if the deviation value is greater than this baseline, its overall temperature level or local temperature peak value deviates from the normal range, and the string is defined as a hot spot string; if the deviation value is much lower than the baseline, it indicates that its temperature is relatively stable and can bear more heat dissipation margin, and it is marked as a cold spot string. Units marked as hot spots usually have positive or significant high temperature deviations in the temperature deviation table. In most cases, the cooling resources of these units are not fully allocated or their internal loads are too heavy. The generation of the distribution map arranges all strings in spatial topology or numbering order, highlights the hot spots with different colors or symbols, and the cold spots also have special marks, so that the distribution of high temperature and low temperature areas can be intuitively distinguished on the map. The graph can also embed gradient information in the temperature state matrix, allowing users to observe the heat flow in two-dimensional or three-dimensional coordinates. In order to ensure the accuracy of the judgment, it is necessary to periodically update the deviation value and recalculate the global average value so that the hot and cold spot marks are always in line with the latest heat dissipation effect. After generating the temperature distribution map, the system will store the map in the memory or database and can be used for the next power transfer operation. Especially when multiple hot spots are concentrated in a certain area, the distribution map can characterize the local heat dissipation imbalance and prompt the corresponding power allocation measures.

[0037] According to the cooling resource allocation, the process of calculating the amount of power that can be reduced by the hot spot string and the amount of power that can be increased by the cold spot string to obtain the maximum allowable power transfer will call the aforementioned generated temperature distribution map and extract the current cooling resource boundary from the dynamic cooling execution instruction set. The hot spot string usually needs to reduce power output or reduce current load under high temperature to alleviate internal heating; the cold spot string has a higher heat dissipation margin and can withstand a certain amount of power increase. The specific implementation often uses a power margin model, which will refer to the temperature deviation value and structural level of the hot spot string, and make a deduction upper limit for the current peak value of its operating power, indicating how many kilowatts or amperes can be reduced without triggering safety hazards. Similarly, the power configuration of the cold spot string will be combined with its low temperature status and the module safety range to calculate the rated upper limit that can be increased to form a comparison table of power that can be increased. The system then matches the power deductions and increase quotas of these hot spots and cold spots one by one, and obtains the maximum allowable power transfer through cumulative calculation. This transfer amount neither breaks the total output balance nor allows the cold spot string to exceed its own heat dissipation capacity. If the cooling resources are still sufficient, the system will allow a wider range of transfers; if the resources are in a tight state, the transfer amount will be compressed to a certain extent. A list of hot spot string power load reduction and a list of cold spot string power expansion can be established in the data structure, and the two can be cross-compared to find a feasible matching point. After each hot spot string completes a transfer protocol with the corresponding cold spot string, the remaining transferable space is updated. This operation ensures that local hot spots can quickly get load reduction opportunities, and cold spots are appropriately loaded, so that the overall power output is still maintained at the required level.

[0038] The power gradient transfer scheme is determined according to the maximum allowable power transfer amount, and the hysteresis control parameters are set for the power transfer timing table in the gradient transfer scheme. In the process of obtaining the power allocation execution matrix, the system will first make a transfer gradient plan based on the maximum allowable power transfer amount, and allocate the load reduction of the hot spot string from the most urgent unit to the cold spot string with the greatest ability to increase power. In order to ensure the smooth power scheduling process, the step-by-step down and step-by-step up methods are adopted. In each period, a part of the hot spot string can be reduced in load, and the other part of the cold spot string can be synchronously increased in output. When constructing the transfer timing table, the algorithm allocates a short delay interval according to the number of strings and the differentiated distance, so that the load change will not instantly impact the electrical balance of the entire system. After that, the hysteresis control parameter setting is applied to the timing table, and an upper and lower threshold is used to determine when to trigger the next stage transfer and when to maintain the current allocation plan. The introduction of hysteresis is mainly to avoid repeated oscillations or frequent small adjustments, so that the load adjustment presents a more continuous behavior. If the temperature fluctuates greatly in a short period of time, the hysteresis can delay the occurrence of frequent switching and reduce the impact on the inverter and battery. Finally, a power allocation execution matrix is ​​formed, which lists the target power range of each string in each time period, and is accompanied by auxiliary information such as transfer sequence number and hysteresis limit. When executing the matrix, the system will implement power regulation section by section, and repeat the same process in the next round of temperature measurement and cooling allocation update. The matrix formed in this way can take into account both real-time cooling and overall grid needs, prevent any hot spot string from being overheated for a long time, and make full use of the available margin of the cold spot string.

[0039] Furthermore, the method of calculating the amount of power that can be reduced by the hot spot string and the amount of power that can be increased by the cold spot string according to the cooling resource allocation to obtain the maximum allowable power transfer includes: obtaining the cooling resource allocation data and actual cooling capacity data of each string in the dynamic cooling execution instruction set to obtain a string cooling efficiency table; calculating the maximum amount of power that can be reduced for each hot spot string according to the string cooling efficiency table and the temperature deviation value of each hot spot string to obtain a hot spot power adjustable scale; calculating the maximum amount of power that can be increased for each cold spot string based on the temperature margin and power carrying capacity of each cold spot string to obtain a cold spot power adjustable scale; performing matching analysis on the hot spot power adjustable scale and the cold spot power adjustable scale to obtain an initial power transfer matching matrix; performing constraint check according to the power transfer direction and value in the initial power transfer matching matrix to obtain the maximum allowable power transfer.

[0040] 104. Select a corresponding temperature control strategy based on the environmental state parameters of the liquid-cooled string-type PCS energy storage inverter and the power allocation execution matrix, wherein the temperature control strategy is used to set parameters of various components of the cooling system corresponding to the liquid-cooled string-type PCS energy storage inverter to achieve temperature management of the liquid-cooled string-type PCS energy storage inverter.

[0041] In one embodiment of the present invention, the method of selecting a corresponding temperature control strategy based on the environmental state parameters of the liquid-cooled string PCS energy storage inverter and the power allocation execution matrix includes: collecting ambient temperature, humidity and prediction data to obtain an environmental state parameter set, and combining the environmental state parameter set with the power allocation execution matrix to perform operating mode identification to obtain the current system operating condition type; matching the corresponding basic strategy group from a preset temperature control strategy library according to the current system operating condition type to obtain a candidate temperature control strategy set; calculating and evaluating the cooling efficiency ratio of each strategy in the candidate temperature control strategy set to obtain a strategy scoring result; selecting a corresponding temperature control strategy based on the strategy scoring result, and generating control parameters for each component of the cooling system according to the temperature control strategy.

[0042] In this embodiment, the temperature data of each string in the liquid-cooled string-type PCS energy storage converter is collected and the temperature characteristic parameters are calculated to obtain a string temperature state matrix; the cooling priority of each string is divided according to the string temperature state matrix and cooling resources are allocated to obtain a dynamic cooling execution instruction set; based on the dynamic cooling execution instruction set and the string temperature state matrix, the hot spot strings and cold spot strings are identified and the power adjustment plan is calculated to obtain a power allocation execution matrix; the corresponding temperature control strategy is selected according to the environmental state parameters and the power allocation execution matrix, and the parameters of each component of the cooling system are set. The present invention combines dynamic cooling resource allocation, power dynamic balance and environmental adaptability temperature control strategy to solve the problems of unreasonable cooling resource allocation and low temperature management efficiency, and realizes accurate temperature management of liquid-cooled string-type PCS energy storage converters under various working conditions.

[0043] The above describes the temperature management method of the liquid-cooled string PCS energy storage converter in the embodiment of the present invention. The following describes the temperature management system of the liquid-cooled string PCS energy storage converter in the embodiment of the present invention. Figure 2 , an embodiment of the temperature management system of the liquid-cooled string PCS energy storage converter in the embodiment of the present invention includes: The temperature monitoring module 201 is used to collect temperature data of each string in the liquid-cooled string-type PCS energy storage converter, and calculate the temperature characteristic parameters based on the temperature data to obtain a string temperature state matrix; A resource allocation module 202, configured to divide the cooling priority of each string according to the string temperature state matrix, and allocate cooling resources according to the cooling priority to obtain a dynamic cooling execution instruction set; A power balancing module 203 is used to identify a string temperature distribution diagram including hot-spot strings and cold-spot strings based on the dynamic cooling execution instruction set and the string temperature state matrix, and calculate a power adjustment scheme according to the string temperature distribution diagram to obtain a power allocation execution matrix; The strategy execution module 204 is used to select a corresponding temperature control strategy according to the environmental state parameters of the liquid-cooled string-type PCS energy storage inverter and the power allocation execution matrix, wherein the temperature control strategy is used to set parameters of each component of the cooling system corresponding to the liquid-cooled string-type PCS energy storage inverter to achieve temperature management of the liquid-cooled string-type PCS energy storage inverter.

[0044] In an embodiment of the present invention, the temperature management system of the liquid-cooled string PCS energy storage converter runs the temperature management method of the liquid-cooled string PCS energy storage converter. The temperature management system of the liquid-cooled string PCS energy storage converter collects temperature data of each string in the liquid-cooled string PCS energy storage converter and calculates temperature characteristic parameters to obtain a string temperature state matrix; divides the cooling priority of each string according to the string temperature state matrix and performs cooling resource allocation to obtain a dynamic cooling execution instruction set; identifies hot spot strings and cold spot strings based on the dynamic cooling execution instruction set and the string temperature state matrix and calculates the power adjustment scheme to obtain a power allocation execution matrix; selects the corresponding temperature control strategy according to the environmental state parameters and the power allocation execution matrix, and sets parameters for each component of the cooling system. The present invention combines dynamic cooling resource allocation, power dynamic balance and environmental adaptability temperature control strategy to solve the problems of unreasonable cooling resource allocation and low temperature management efficiency, and realizes precise temperature management of liquid-cooled string PCS energy storage converters under various working conditions.

[0045] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0046] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0047] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A temperature management method for a liquid-cooled string PCS energy storage converter, characterized in that: The temperature management method of the liquid-cooled string PCS energy storage converter includes: Collecting temperature data of each string in the liquid-cooled string-type PCS energy storage converter, and calculating temperature characteristic parameters based on the temperature data to obtain a string temperature state matrix; Dividing the cooling priority of each string according to the string temperature state matrix, and allocating cooling resources according to the cooling priority to obtain a dynamic cooling execution instruction set; Based on the dynamic cooling execution instruction set and the string temperature state matrix, a string temperature distribution diagram including hot spot strings and cold spot strings is identified, and a power adjustment scheme is calculated according to the string temperature distribution diagram to obtain a power allocation execution matrix; A corresponding temperature control strategy is selected according to the environmental state parameters of the liquid-cooled string-type PCS energy storage inverter and the power allocation execution matrix, wherein the temperature control strategy is used to set parameters of various components of the cooling system corresponding to the liquid-cooled string-type PCS energy storage inverter to achieve temperature management of the liquid-cooled string-type PCS energy storage inverter.

2. The temperature management method of the liquid-cooled string PCS energy storage converter according to claim 1 is characterized in that: The step of collecting temperature data of each string in the liquid-cooled string-type PCS energy storage converter and calculating temperature characteristic parameters based on the temperature data to obtain a string temperature state matrix includes: The temperature data is collected by using a temperature sensor array arranged at key points of each PCS module and battery cluster of the liquid-cooled string PCS energy storage converter to obtain an original temperature data set; The temperature original data set is differentiated based on the current power state and position information of each string to obtain a string temperature threshold setting matrix; Calculate the difference between the current temperature of each string and the corresponding threshold in the string temperature threshold setting matrix to obtain the temperature safety margin of each string; The temperature difference between adjacent strings and the ratio of the physical distance are calculated to obtain the temperature gradient vector between the strings; The temperature safety margin and the temperature gradient vector are organized in a matrix according to the string numbers to obtain a string temperature state matrix.

3. The temperature management method of the liquid-cooled string PCS energy storage converter according to claim 1, characterized in that: The step of dividing the cooling priority of each string according to the string temperature state matrix, and allocating cooling resources according to the cooling priority to obtain a dynamic cooling execution instruction set includes: According to the temperature safety margin in the string temperature state matrix, each string is divided into three levels of cooling priority to obtain a string cooling priority classification table; Based on the string cooling priority classification table, the speed of the main circulation pump of the cooling system of the liquid-cooled string PCS energy storage converter is adjusted and calculated to obtain the main circulation flow control parameter; Calculate the flow distribution ratio of the coolant pipeline section corresponding to each string according to the main circulation flow control parameter and the string cooling priority classification table to obtain a group pipeline control instruction; Calculate the microchannel flow adjustment scheme inside each PCS module based on the grouped pipeline control instructions and the internal heat distribution data of each string to obtain a microchannel control instruction; The main circulation flow control parameters, the grouped pipeline control instructions and the microchannel control instructions are collaboratively optimized to obtain a dynamic cooling execution instruction set.

4. The temperature management method of the liquid-cooled string PCS energy storage converter according to claim 3 is characterized in that: The main circulation flow control parameters obtained by adjusting and calculating the speed of the main circulation pump of the cooling system of the liquid-cooled string PCS energy storage converter based on the string cooling priority classification table include: Performing statistical analysis on the number and temperature safety margin of each priority string in the string cooling priority classification table to obtain the overall cooling demand value of the system; According to the overall cooling demand value of the system, a preset pump speed-flow mapping table is queried to obtain a basic main circulation pump speed value; When the proportion of high priority strings exceeds a threshold, the basic main circulation pump speed value is corrected for emergency cooling demand to obtain a corrected pump speed value; The pump power control curve is calculated based on the corrected pump speed value combined with the system pipeline pressure data to obtain a pump speed control sequence, and the pump speed control sequence is converted into a pulse width modulation signal parameter recognized by the pump controller to obtain the main circulation flow control parameter.

5. The temperature management method of the liquid-cooled string PCS energy storage converter according to claim 1, characterized in that: The identifying of a string temperature distribution diagram including hot spot strings and cold spot strings based on the dynamic cooling execution instruction set and the string temperature state matrix, and calculating a power adjustment scheme according to the string temperature distribution diagram to obtain a power allocation execution matrix includes: Calculate the temperature deviation value of each string according to the string temperature state matrix and the cooling resource allocation in the dynamic cooling execution instruction set to obtain a string temperature deviation table; Based on the string temperature deviation table, the strings with a temperature higher than the average value are marked as hot-spot strings, and the strings with a temperature lower than the average value are marked as cold-spot strings, so as to obtain a string temperature distribution diagram; According to the cooling resource allocation, the amount of power that can be reduced by the hot spot group string and the amount of power that can be increased by the cold spot group string are calculated to obtain the maximum allowable power transfer amount; A power gradient transfer scheme is determined according to the maximum allowable power transfer amount, and hysteresis control parameters are set for the power transfer timing table in the gradient transfer scheme to obtain a power allocation execution matrix.

6. The temperature management method of the liquid-cooled string PCS energy storage converter according to claim 5, characterized in that: The calculating, according to the cooling resource allocation, the amount of power that can be reduced by the hot spot group string and the amount of power that can be increased by the cold spot group string to obtain the maximum allowable power transfer amount comprises: Acquire cooling resource allocation data and actual cooling capacity data of each string in the dynamic cooling execution instruction set to obtain a string cooling efficiency table; Calculate the maximum reducible power of each hotspot string according to the string cooling efficiency table and the temperature deviation value of each hotspot string, and obtain the hotspot power adjustable table; Based on the temperature margin and power carrying capacity of each cold spot string, the maximum power that can be increased for each cold spot string is calculated to obtain an adjustable cold spot power scale; Performing matching analysis on the hot spot power adjustable scale and the cold spot power adjustable scale to obtain an initial power transfer matching matrix; A constraint condition check is performed according to the power transfer direction and value in the initial power transfer matching matrix to obtain a maximum allowable power transfer amount.

7. The temperature management method of the liquid-cooled string PCS energy storage converter according to claim 1, characterized in that: The selecting of the corresponding temperature control strategy according to the environmental state parameters of the liquid-cooled string PCS energy storage converter and the power allocation execution matrix includes: Collecting ambient temperature, humidity and prediction data to obtain an ambient state parameter set, and combining the ambient state parameter set with the power allocation execution matrix to perform operation mode identification to obtain the current system operating condition type; According to the current system operating condition type, a corresponding basic strategy group is matched from a preset temperature control strategy library to obtain a candidate temperature control strategy set; Calculating and evaluating the cooling efficiency ratio of each strategy in the candidate temperature control strategy set to obtain a strategy scoring result; A corresponding temperature control strategy is selected based on the strategy scoring result, and control parameters of various components of the cooling system are generated according to the temperature control strategy.

8. A temperature management system for a liquid-cooled string PCS energy storage converter, characterized in that: The temperature management system of the liquid-cooled string PCS energy storage converter includes: A temperature monitoring module is used to collect temperature data of each string in the liquid-cooled string-type PCS energy storage converter, and calculate temperature characteristic parameters based on the temperature data to obtain a string temperature state matrix; A resource allocation module, used to divide the cooling priority of each string according to the string temperature state matrix, and allocate cooling resources according to the cooling priority to obtain a dynamic cooling execution instruction set; A power balancing module, configured to identify a string temperature distribution diagram including hot-spot strings and cold-spot strings based on the dynamic cooling execution instruction set and the string temperature state matrix, and calculate a power adjustment scheme according to the string temperature distribution diagram to obtain a power allocation execution matrix; A strategy execution module is used to select a corresponding temperature control strategy according to the environmental state parameters of the liquid-cooled string-type PCS energy storage inverter and the power allocation execution matrix, wherein the temperature control strategy is used to set parameters of various components of the cooling system corresponding to the liquid-cooled string-type PCS energy storage inverter to achieve temperature management of the liquid-cooled string-type PCS energy storage inverter.

Images