Cooling cooperative control method and device, transformer and storage medium

Abstract

The invention discloses a cooling cooperative control method, and the method comprises the steps: obtaining a temperature parameter, a load state parameter and a cooling system parameter of a transformer, and forming a parameter set; constructing a gating circulation unit model, and predicting thermal load distribution of each region of the transformer by adopting the parameter set through the gating circulation unit model; and performing real-time distribution on cooling resources by adopting a dynamic region distribution algorithm. According to the cooling cooperative control method, temperature and load changes of all areas of the transformer are predicted in real time through a gating circulation unit model; intelligent distribution of cooling resources is achieved through a dynamic area distribution algorithm, the cooling cooperative control method can dynamically adjust the flow speed of a cooling medium and the rotating speed of a fan according to real-time temperature and load changes, it is ensured that all areas of a transformer obtain appropriate cooling resources, and extra cooling is preferentially provided for areas with the temperature abnormally increased; and a cooling strategy is optimized through multiple iterations in combination with historical operation data, so that the response speed and accuracy of the cooling system are remarkably improved.

Description

Technical Field

[0001] The present application relates to the technical field of equipment cooling, and in particular to a cooling collaborative control method, device and storage medium. Background Art

[0002] In transformers, current cooling systems are mostly based on the overall load or temperature of the transformer, using a static cooling mode. However, during operation, transformers often have uneven load distribution and different electrical spacing. Some local areas may overheat due to excessive load or too small spacing. Traditional cooling systems cannot accurately sense and quickly respond to these local overheating, which may cause the temperature of local areas to continue to rise, thus affecting the safe operation of the transformer.

[0003] At the same time, the regulation process of conventional cooling systems is usually relatively fixed and cannot be dynamically adjusted according to the load and temperature changes during transformer operation. Once the cooling equipment is started, it usually works according to the preset time and power. Even if the local load decreases or the temperature drops, the cooling system will not adjust immediately, resulting in a waste of cooling resources and a decrease in cooling efficiency. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a cooling collaborative control method that can accurately sense and respond to local overheating in real time through a gated cycle unit model and a dynamic area allocation algorithm.

[0005] According to the cooling collaborative control method of the first aspect of the present application, the cooling collaborative control method includes:

[0006] Obtain the temperature parameters, load state parameters and cooling system parameters of the transformer to form a parameter set;

[0007] A gated cyclic unit model is constructed, and the thermal load distribution of each area of ​​the transformer is predicted by using the parameter set through the gated cyclic unit model;

[0008] According to the prediction results of the gated recurrent unit model, a dynamic area allocation algorithm is used to allocate cooling resources in real time.

[0009] According to the cooling collaborative control method of the embodiment of the present application, there are at least the following beneficial effects: the cooling collaborative control method introduces a gated loop unit model, and uses the gated loop unit model to predict the temperature and load changes of each area of ​​the transformer in real time; a dynamic area allocation algorithm is used to realize the intelligent allocation of cooling resources. Compared with the traditional static cooling mode, the cooling collaborative control method can dynamically adjust the cooling medium flow rate and fan speed according to the real-time temperature and load changes to ensure that each area of ​​the transformer obtains appropriate cooling resources, and provides additional cooling for areas with abnormally high temperatures to avoid local overheating problems. The cooling strategy is optimized through multiple iterations based on historical operation data, which significantly improves the response speed and accuracy of the cooling system.

[0010] According to some embodiments of the present application, a gated cyclic unit model is constructed, and a parameter set is used to predict the thermal load distribution of each area of ​​the transformer through the gated cyclic unit model, including:

[0011] Build a gated recurrent unit model including hidden states based on the PyTorch framework;

[0012] The parameter set is input into the gated recurrent unit model, and the predicted temperature value of each region at a future time is output according to the hidden state.

[0013] According to some embodiments of the present application, after the parameter set is input into the gated recurrent unit model and the temperature prediction value of each region at a future time is output according to the hidden state, it includes:

[0014] Based on the output results of the gated cyclic unit model and the change trend of the load state parameters, the thermal load prediction value of each area of ​​the transformer is obtained;

[0015] By comprehensively evaluating the heat load prediction value and the temperature prediction value, a local overheating area risk function is constructed.

[0016] According to some embodiments of the present application, a dynamic area allocation algorithm is used to allocate cooling resources in real time according to the prediction results of the gated cycle unit model, including:

[0017] Construct cooling resource allocation function based on temperature prediction value and heat load prediction value;

[0018] Dynamically adjust the cooling medium flow rate and fan speed in each area of ​​the transformer according to the cooling resource allocation function;

[0019] When the temperature prediction value of a specific area exceeds the preset temperature threshold, the cooling resource allocation amount of the area is increased.

[0020] According to some embodiments of the present application, after allocating cooling resources in real time using a dynamic area allocation algorithm according to the prediction result of the gated cycle unit model, the method includes:

[0021] Adopting an adaptive algorithm based on proximal strategy optimization to automatically learn the relationship between transformer load change and temperature change;

[0022] Each cooling unit of the transformer makes cooling decisions independently based on local temperature and load information, and achieves collaborative control by sharing global information.

[0023] According to some embodiments of the present application, an adaptive algorithm based on proximal strategy optimization is used to automatically learn the relationship between transformer load change and temperature change, including:

[0024] Construct a state set based on the real-time load change and temperature change of the transformer;

[0025] According to the adaptive algorithm optimized by the proximal strategy, a cooling resource allocation strategy is constructed for the state set, and it is optimized iteratively;

[0026] The updated cooling resource allocation strategy is used to calculate the cooling resource allocation amount for each area of ​​the transformer at the next moment.

[0027] According to some embodiments of the present application, each cooling unit of the transformer independently makes a cooling decision based on local temperature and load information, and realizes collaborative control by sharing global information, including:

[0028] Each cooling unit of the transformer makes independent cooling decisions based on local temperature and local load information, and independently calculates the cooling resource allocation;

[0029] Global information sharing among transformer cooling units to build a global state set;

[0030] The cooling resources of each cooling unit of the transformer are globally optimized through a collaborative control mechanism, and the global cooling resource allocation results are fed back to each cooling unit.

[0031] According to some embodiments of the present application, each cooling unit of the transformer independently makes a cooling decision according to local temperature and load information, and realizes collaborative control by sharing global information, including:

[0032] Monitor the power consumption and cooling efficiency of each cooling unit of the transformer, and dynamically adjust the working status of each cooling unit based on the real-time load of the transformer;

[0033] When abnormal temperature rise or overheating risk occurs in a specific area of ​​the transformer, the system automatically triggers cooling adjustments through the loop.

[0034] According to the cooling collaborative control device of the second aspect of the embodiment of the present application, the cooling collaborative control device includes a data acquisition module, a processing module, an energy consumption optimization module and a monitoring and feedback module. The data acquisition module is used to collect and obtain parameter sets; the processing module is used to predict the thermal load distribution of each area of ​​the transformer and perform real-time allocation of cooling resources; the energy consumption optimization module is used to monitor the power consumption and cooling efficiency of each cooling unit; the monitoring and feedback module is used to identify abnormal temperature rise or overheating risks and trigger cooling adjustments.

[0035] According to the transformer of the third aspect embodiment of the present application, the transformer includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the program to implement the above-mentioned cooling coordinated control method.

[0036] According to the computer-readable storage medium of the fourth aspect embodiment of the present application, the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the above-mentioned cooling coordinated control method.

[0037] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings are used to provide a further understanding of the technical solution disclosed in the present application and constitute a part of the specification. Together with the embodiments disclosed in the present application, they are used to explain the technical solution disclosed in the present application and do not constitute a limitation on the technical solution disclosed in the present application.

[0039] Figure 1 This is a flowchart of a cooling collaborative control method provided by an embodiment of the present application.

[0040] Figure 2 This is a flowchart of a cooling collaborative control method provided by another embodiment of the present application.

[0041] Figure 3 is a flow chart of a cooling collaborative control method provided by another embodiment of the present application;

[0042] Figure 4 is a flow chart of a cooling collaborative control method provided by another embodiment of the present application;

[0043] Figure 5 is a flow chart of a cooling collaborative control method provided by another embodiment of the present application;

[0044] Figure 6 is a flow chart of a cooling collaborative control method provided by another embodiment of the present application;

[0045] Figure 7is a flow chart of a cooling collaborative control method provided by another embodiment of the present application;

[0046] Figure 8 is a flow chart of a cooling collaborative control method provided by another embodiment of the present application;

[0047] Fig. 9 It is an architectural diagram of a cooling system involved in a cooling collaborative control method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0049] In the description of the present application, it should be understood that descriptions involving orientation, such as up, down, front, back, left, right, etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0050] In the description of this application, "several" means more than one, "more" means more than two, "greater than", "less than", "exceed", etc. are understood to exclude the number itself, and "above", "below", "within", etc. are understood to include the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0051] In the description of this application, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in this application based on the specific content of the technical solution.

[0052] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0053] The embodiment of the present application provides a cooling collaborative control method, which is applied to a transformer, specifically, a box-type transformer. The box-type transformer has several areas, each area is provided with several independent cooling units, and each cooling unit is used to cool the transformer.

[0054] In the cooling collaborative control method, a gated cyclic unit model is used to predict the temperature and load changes in each area of ​​the transformer. At the same time, a dynamic area allocation algorithm is used to realize the intelligent allocation of cooling resources, reduce energy consumption, and improve the reliability and cooling efficiency of transformer operation.

[0055] like Figure 1 As shown, Figure 1 It is a flowchart of a cooling collaborative control method provided by an embodiment of the present application. The cooling collaborative control method includes but is not limited to step S110, step S120, and step S130.

[0056] Step S110, obtaining temperature parameters, load state parameters and cooling system parameters of the transformer to form a parameter set.

[0057] In some examples, the temperature parameters T of each area of ​​the transformer are collected in real time. i (t); Real-time collection of load state parameters L in each area of ​​the transformer i (t); Real-time collection of cooling unit parameters C of transformer cooling system j (t), this parameter includes the flow rate of the cooling medium and the speed of the fan.

[0058] Furthermore, the operating parameter set P(t) is constructed based on all the above parameters:

[0059] P(t)={T i (t),L i (t),C j (t)∣i=1,2,...,n; j=1,2,...,m};

[0060] Wherein, n is the total number of regions of the transformer, m is the total number of cooling units, i represents the i-th region, t represents time, and j represents the j-th cooling unit.

[0061] Step S120: constructing a gated cyclic unit model, and using the gated cyclic unit model to predict the thermal load distribution of each region of the transformer using a parameter set.

[0062] Step S130, according to the prediction result of the gated cycle unit model, a dynamic area allocation algorithm is used to allocate cooling resources in real time.

[0063] In addition, refer to Figure 2 In some examples, step S120 may include but is not limited to the following steps:

[0064] Step S210, constructing a gated recurrent unit model including a hidden state based on the PyTorch framework.

[0065] In some examples, a system for processing the transformer historical operating parameter set P is constructed based on the PyTorch framework. hist (t), the updating process of the gated recurrent unit model includes:

[0066] z t =σ(W z ·[P hist (t),h t-1 ]+b z );

[0067] r t =σ(W r ·[P hist (t),h t-1 ]+b r );

[0068] h' t =tanh(W h ·[P hist (t),r t oh t-1 ]+b h );

[0069] h t =(1-z t )oh t-1 +z t oh' t ;

[0070] Among them, z t is the update gate, r t is the reset gate, h t is the hidden state, P hist (t) is the historical operating parameter set input, W z , W r , W h are weight matrices, b z 、b r 、b h are biases, σ is the activation function, and o is the element-by-element product operation.

[0071] Step S220, input the parameter set into the gated recurrent unit model, and output the predicted temperature value of each region at a future time according to the hidden state.

[0072] In some examples, the historical operating parameter set P hist (t) Input to the gated recurrent unit model, based on the hidden state ht Output the predicted temperature value of each area at the future time

[0073]

[0074] in, represents the predicted temperature value of the i-th region at the future time t+k, W T is the weight matrix for temperature prediction, b T Bias for temperature prediction.

[0075] It is understandable that the cooling collaborative control method solves the problem that traditional transformer cooling systems cannot accurately deal with local overheating. By collecting data on real-time temperature and load parameters in each area of ​​the transformer and using the gated recurrent unit (GRU) model to analyze the historical operating parameter set, the solution can predict the temperature changes and local heat load distribution of each area in the future in advance, providing forward-looking overheating risk identification.

[0076] This prediction process realizes intelligent perception of local areas, can identify possible local overheating before the system is running, and improve the response speed and accuracy of the cooling system. This method provides a basis for subsequent cooling resource allocation and realizes pre-scheduling of cooling resources based on load prediction in the transformer field.

[0077] In addition, refer to Figure 3 In some examples, step S220 may include but is not limited to the following steps:

[0078] Step S230 , based on the output result of the gated cyclic unit model and the change trend of the load state parameter, the thermal load prediction value of each area of ​​the transformer is obtained.

[0079] In some examples, the output of the gated recurrent unit model is combined with the load state parameter L i (t) to predict the heat load distribution in each area and calculate the predicted heat load value of each area.

[0080]

[0081] Among them, α, β, and γ are weight coefficients based on historical data regression. is the predicted heat load value of the i-th area at the future time t+k, ΔL i (t) indicates the current load state change, C j (t) is the cooling unit status.

[0082] Step S240, constructing a local overheating area risk function by comprehensively evaluating the heat load prediction value and the temperature prediction value.

[0083] In some examples, the heat load prediction value is With temperature prediction value Conduct comprehensive assessment and construct local overheating area risk function hot (t+k):

[0084]

[0085] Among them, L max is the maximum allowable load, T max is the maximum allowable temperature, O hot (t+k) represents the local overheating risk value at time t+k.

[0086] In addition, refer to Figure 4 In some examples, step S130 may include but is not limited to the following steps:

[0087] Step S310: construct a cooling resource allocation function according to the temperature prediction value and the heat load prediction value.

[0088] In some examples, the temperature prediction value based on the gated recurrent unit model and heat load forecast Construct cooling resource allocation function R i (t+k), which is used to calculate the amount of cooling resources required for the i-th zone:

[0089]

[0090] Among them, λ1 and λ2 are weight coefficients based on historical data fitting, R i (t+k) represents the amount of cooling resources required by the i-th area at the future time t+k.

[0091] It is understandable that in view of the difference in heat load in different areas of the transformer, traditional cooling systems usually use fixed thresholds for cooling adjustment, resulting in resource waste. The dynamic regional allocation algorithm proposed in this method realizes the intelligent optimization allocation of cooling resources based on regional predicted heat load. Through the cooling resource allocation function:

[0092]

[0093] This method dynamically adjusts the cooling medium flow rate and fan speed according to the predicted temperature and load values ​​of different areas, and gives priority to providing additional cooling resources for high-temperature areas. This dynamic cooling resource allocation mechanism significantly improves cooling efficiency, breaks the limitations of fixed cooling thresholds, and can be flexibly adjusted according to real-time temperature changes in local areas. While effectively reducing energy consumption, it ensures balanced cooling of the system in each load area and avoids excessive consumption of cooling resources. Compared with the traditional "global" cooling strategy, this partition-optimized cooling strategy is extremely innovative in the complex environment of transformers.

[0094] Step S320: dynamically adjust the cooling medium flow rate and fan speed of each area of ​​the transformer according to the cooling resource allocation function.

[0095] In some examples, according to the cooling resource allocation function R i (t+k), dynamically adjust the cooling medium flow rate F in each area i (t+k) and fan speed V i (t+k), the cooling medium flow rate and fan speed are adjusted as follows:

[0096] F i (t+k)=F base +δ F ·R i (t+k);

[0097] V i (t+k)=V base +δ V ·R i (t+k);

[0098] Among them, F base and V base Based on coolant / oil flow rate and fan speed, δ F and δ V Incremental adjustment coefficients control the dynamic adjustment of coolant / oil flow rate and fan speed respectively.

[0099] Step S330: when the temperature prediction value of a specific area exceeds a preset temperature threshold, increase the cooling resource allocation amount of the area.

[0100] In some examples, additional cooling resources are provided preferentially to high-temperature areas. Exceeds the preset temperature threshold T threshold Increase its cooling resource allocation when:

[0101]

[0102] Where η is the allocation increment coefficient of additional cooling resources, R' i(t+k) is the adjusted cooling resource allocation.

[0103] Furthermore, the cooling resource allocation R' of each area of ​​the transformer is i (t+k) is transmitted to the cooling control system, which controls each cooling unit to perform corresponding cooling operations according to the cooling resource allocation, so that the high-temperature area is cooled preferentially.

[0104] In addition, refer to Figure 5 In some examples, step S130 may include but is not limited to the following steps:

[0105] Step S140: automatically learning the relationship between the transformer load change and the temperature change using an adaptive algorithm based on proximal strategy optimization.

[0106] In step S150, each cooling unit of the transformer independently makes a cooling decision according to local temperature and load information, and realizes collaborative control by sharing global information.

[0107] In addition, refer to Figure 6 In some examples, step S140 may include but is not limited to the following steps:

[0108] Step S410: construct a state set according to the real-time load change and temperature change of the transformer.

[0109] In some examples, the real-time load change L of the transformer is obtained. i (t) and temperature change T i (t), and use this to construct the state set S(t), which includes the current load and temperature conditions:

[0110] S(t)={L i (t),T i (t)|i=1,2,...,p};

[0111] Where p is the total number of regions of the transformer, and i represents the i-th region.

[0112] Step S420: construct a cooling resource allocation strategy for the state set according to the adaptive algorithm of proximal strategy optimization, and perform iterative optimization.

[0113] In some examples, a cooling resource allocation strategy π is constructed for the state set S(t) using an adaptive algorithm for proximal policy optimization. θ (a|S(t)), a is the action of cooling resource allocation, θ is the strategy parameter, and the cooling resource allocation strategy is updated according to the following formula:

[0114]

[0115] Among them, α is the learning rate, is the advantage function, which is used to measure the pros and cons of actions under the current strategy.

[0116] It is understandable that most traditional cooling systems lack intelligent cooling resource scheduling mechanisms, resulting in delayed response and waste of resources. This method introduces an adaptive proximal policy optimization algorithm (PPO) to achieve adaptive learning of the cooling system to transformer load and temperature changes. Through the dynamic update of the state set S(t) and the intelligent learning of the cooling strategy, the system can continuously optimize the cooling resource allocation strategy through multiple iterations, adjust the cooling medium flow rate and fan speed to maximize the cooling effect and minimize energy consumption:

[0117]

[0118] Among them, A t is an advantage function used to measure the pros and cons of actions under the current strategy. Through this intelligent cooling scheduling mechanism, the system can respond quickly to load changes and automatically adjust the allocation of cooling resources without relying on manual intervention, thereby greatly improving the flexibility and intelligence of the system. In application scenarios with frequent load fluctuations, this solution shows significant technical gains and reduces the lag problem of cooling response.

[0119] Step S430: using the updated cooling resource allocation strategy to calculate the cooling resource allocation amount for each area of ​​the transformer at the next moment.

[0120] In some examples, the system optimizes the cooling resource allocation strategy based on multiple iterations of the proximal strategy optimization algorithm, learns the relationship between load changes and temperature changes, and then automatically adjusts the cooling resource allocation in each area of ​​the transformer to maximize the cooling effect and minimize energy consumption.

[0121] Furthermore, in each iteration, the updated cooling resource allocation strategy π is used θ (a|S(t)) calculates the cooling resource allocation R of each area of ​​the transformer at the next moment i (t+1):

[0122]

[0123] Among them, β is the update step size, R i (t+1) is the cooling resource allocation amount of the i-th area at the next moment.

[0124] In addition, refer to Figure 7 In some examples, step S150 may include but is not limited to the following steps:

[0125] In step S510, each cooling unit of the transformer independently makes a cooling decision according to local temperature and local load information, and independently calculates the cooling resource allocation amount.

[0126] In some examples, each cooling unit in the transformer is treated as an independent agent, Agent A j Based on the local temperature T j (t) and local load information L j (t) Make cooling decisions independently. For the jth cooling unit, the cooling decision strategy π j It is expressed as:

[0127] π j (a j |S j (t)) = f j (T j (t),L j (t));

[0128] Among them, a j is the decision action of the cooling unit, S j (t) is the state set of the jth cooling unit, including temperature and load information.

[0129] Furthermore, each cooling unit independently calculates the cooling resource allocation R according to the local temperature and load information. j (t):

[0130] R j (t) = μ1·T j (t)+μ2·L j (t);

[0131] Among them, μ1 and μ2 are the weight coefficients of local cooling resource allocation, R j (t) is the cooling resource allocation of the jth cooling unit.

[0132] Step S520: Global information is shared among cooling units of the transformer to construct a global state set.

[0133] In some examples, the global information sharing among cooling units is realized by using the Internet of Things technology to build a global state set S global (t), the global state set includes the temperature and load information of all cooling units:

[0134] S global (t) = {T j (t),L j (t)|j=1,2,...,m};

[0135] Wherein, m is the total number of cooling units.

[0136] Step S530, globally optimizing the cooling resources of each cooling unit of the transformer through a collaborative control mechanism, and feeding back the global cooling resource allocation result to each cooling unit.

[0137] In some examples, in the global state set S global (t), the cooling resources of each cooling unit of the transformer are globally optimized through a collaborative control mechanism, the global cooling resource allocation result is fed back to each cooling unit, and the local cooling strategy of each cooling unit is adjusted according to the global cooling resource allocation result.

[0138] It can be understood that this method treats each cooling unit as an independent intelligent entity, independently makes cooling decisions based on local temperature and load information, and shares global information through Internet of Things technology, so that the system can achieve collaborative optimization of global cooling resources.

[0139] While the cooling unit makes decisions locally, it also adjusts according to the global state set to achieve the optimal allocation of cooling resources globally. Formula:

[0140]

[0141] In addition, refer to Figure 8 In some examples, step S150 may include but is not limited to the following steps:

[0142] Step S160, monitoring the power consumption and cooling efficiency of each cooling unit of the transformer, and dynamically adjusting the working state of each cooling unit in combination with the real-time load condition of the transformer.

[0143] In some examples, the power consumption and cooling efficiency of each cooling unit of the transformer are monitored through the energy consumption optimization module, and the working state of each cooling unit is dynamically adjusted according to the real-time load condition of the transformer, so as to minimize the system energy consumption while ensuring the cooling effect.

[0144] Step S170: When abnormal temperature rise or overheating risk occurs in a specific area of ​​the transformer, the system automatically triggers cooling adjustment through the loop.

[0145] In some examples, first, a remote monitoring system is built to monitor the temperature parameters T of each area of ​​the transformer based on the Internet of Things technology. i (t) and the operating status of the cooling system C j (t) to conduct real-time monitoring and form a complete monitoring data set M(t):

[0146] M(t)={T i (t),C j (t)∣i=1,2,...,n1; j=1,2,...,m1};

[0147] Where n1 is the total number of transformer areas and m1 is the total number of cooling units.

[0148] Secondly, through real-time analysis of the monitoring data set M(t), when the system detects the temperature T of a specific area of ​​the transformer i (t) exceeds the preset threshold T threshold The cooling adjustment mechanism is automatically triggered when the transformer is in a state of low temperature. The cooling adjustment mechanism controls the redistribution of cooling resources through an intelligent feedback loop and makes incremental adjustments to cooling resources for the corresponding areas of the transformer:

[0149] R' i (t) = R i (t)+δ·(T i (t)-T threshold );

[0150] Among them, δ is the cooling resource adjustment coefficient, R' i (t) is the adjusted cooling resource allocation.

[0151] Again, in the feedback loop, the system monitors the temperature changes in a single area of ​​the transformer and integrates the working status of the transformer cooling unit C j (t), so that each cooling unit can work together when the temperature rises abnormally. The system updates the cooling adjustment mechanism according to the real-time updated temperature and cooling unit status data, and dynamically optimizes the cooling resource allocation strategy.

[0152] Then, while the system automatically performs cooling adjustments, the remote monitoring system records and transmits detailed information about each cooling adjustment to the user interface through the data feedback module. Users can view the temperature status of each area of ​​the transformer and the adjustment process of the cooling system in real time through the remote monitoring platform.

[0153] An embodiment of the present application provides a cooling collaborative control device, which includes a data acquisition module, a processing module, an energy consumption optimization module, and a monitoring and feedback module.

[0154] Among them, the data acquisition module is used to collect and obtain parameter sets; the processing module is used to predict the heat load distribution of each area of ​​the transformer and to allocate cooling resources in real time; the energy consumption optimization module is used to monitor the power consumption and cooling efficiency of each cooling unit; the monitoring and feedback module is used to identify abnormal temperature rise or overheating risks and trigger cooling adjustments.

[0155] at the same time, Fig. 9 The present invention is an architectural diagram of a cooling system involved in a cooling collaborative control method. The cooling system includes a cooling system controller. The cooling system controller is electrically connected to a plurality of cooling units. Each cooling unit is located in a respective area of ​​the transformer and is used for cooling work in each area.

[0156] Furthermore, each cooling unit makes cooling decisions based on the local state, thereby forming a global state set, which feeds information back to each cooling unit.

[0157] It can be understood that the cooling collaborative control method introduces a gated loop unit model to make real-time predictions on the temperature and load changes in each area of ​​the box-type transformer, and uses a dynamic area allocation algorithm to achieve intelligent allocation of cooling resources. Compared with the traditional static cooling mode, the present invention can dynamically adjust the cooling medium flow rate and fan speed according to real-time temperature and load changes to ensure that each area obtains appropriate cooling resources, and provides additional cooling priority to areas with abnormally high temperatures to avoid local overheating problems. The cooling strategy is optimized through multiple iterations based on historical operating data, significantly improving the response speed and accuracy of the cooling system.

[0158] The cooling collaborative control method dynamically learns the relationship between transformer load changes and temperature changes through an adaptive learning mechanism based on the proximal strategy optimization algorithm, thereby continuously optimizing the cooling resource allocation strategy during system operation and achieving maximum optimization of energy consumption. Each cooling unit, as an independent intelligent entity, makes independent decisions based on local temperature and load information, and achieves collaborative control by sharing global state information to optimize cooling resource allocation on a global scale. Compared with the phenomenon of overcooling or uneven cooling in traditional cooling systems, this method reduces unnecessary energy consumption and improves overall cooling efficiency while ensuring the safe operation of the transformer.

[0159] The cooling collaborative control method combines the Internet of Things technology to build a remote monitoring and data feedback mechanism, which can monitor the temperature and cooling system status of the box-type transformer in real time. When an abnormal temperature rise occurs in a certain area, the system automatically triggers the cooling adjustment mechanism through an intelligent feedback loop and updates the cooling model in real time to ensure that the system cooling strategy can be continuously optimized with environmental changes and load conditions. At the same time, users can view the temperature status and cooling adjustment process of each area in real time through the remote monitoring platform, which improves the intelligence and response speed of the system and reduces the necessity of manual intervention.

[0160] Specifically, Fig. 9 There are 6 cooling units, namely A1, A2, A3, A4, A5 and A6, 6 local states, namely S1(t), S2(t), S3(t), S4(t), S5(t) and S6(t), and 6 cooling decisions, namely π1, π2, π3, π4, π5 and π6.

[0161] The present application embodiment provides a transformer, which may include:

[0162] A memory, a processor, and a computer program stored in the memory and executable on the processor.

[0163] When the processor executes the program, the cooling coordinated control method provided in the above embodiment is implemented.

[0164] Furthermore, the transformer further comprises:

[0165] Communication interface, used for communication between memory and processor.

[0166] Memory is used to store computer programs that can be run on the processor.

[0167] The memory may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0168] If the memory, processor and communication interface are implemented independently, the communication interface, memory and processor can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc.

[0169] Optionally, in a specific implementation, if the memory, processor and communication interface are integrated on a chip, the memory, processor and communication interface can communicate with each other through an internal interface.

[0170] The processor may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0171] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the above-mentioned cooling coordinated control method is implemented.

[0172] This embodiment also provides a computer program product, including a computer program, which is used to implement the above cooling coordinated control method when executed.

[0173] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0174] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0175] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0176] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, decompressing or processing in other suitable ways if necessary, and then storing it in a computer memory.

[0177] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0178] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0179] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0180] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A cooling coordinated control method, characterized in that: include: Obtain the temperature parameters, load state parameters and cooling system parameters of the transformer to form a parameter set; A gated cyclic unit model is constructed, and the thermal load distribution of each area of ​​the transformer is predicted by using the parameter set through the gated cyclic unit model; According to the prediction results of the gated recurrent unit model, a dynamic area allocation algorithm is used to allocate cooling resources in real time.

2. The cooling coordinated control method according to claim 1, characterized in that: A gated cyclic unit model is constructed to predict the thermal load distribution of each area of ​​the transformer using a parameter set, including: Build a gated recurrent unit model including hidden states based on the PyTorch framework; The parameter set is input into the gated recurrent unit model, and the predicted temperature value of each region at a future time is output according to the hidden state.

3. The cooling coordinated control method according to claim 2, characterized in that: After inputting the parameter set into the gated recurrent unit model and outputting the predicted temperature value of each region at the future time according to the hidden state, it includes: Based on the output results of the gated cyclic unit model and the change trend of the load state parameters, the thermal load prediction value of each area of ​​the transformer is obtained; By comprehensively evaluating the heat load prediction value and the temperature prediction value, a local overheating area risk function is constructed.

4. The cooling coordinated control method according to claim 3, characterized in that: Based on the prediction results of the gated cycle unit model, a dynamic area allocation algorithm is used to allocate cooling resources in real time, including: Construct cooling resource allocation function based on temperature prediction value and heat load prediction value; Dynamically adjust the cooling medium flow rate and fan speed in each area of ​​the transformer according to the cooling resource allocation function; When the temperature prediction value of a specific area exceeds the preset temperature threshold, the cooling resource allocation amount of the area is increased.

5. The cooling coordinated control method according to claim 1, characterized in that: According to the prediction results of the gated cycle unit model, the dynamic area allocation algorithm is used to allocate cooling resources in real time, including: Adopting an adaptive algorithm based on proximal strategy optimization to automatically learn the relationship between transformer load change and temperature change; Each cooling unit of the transformer makes cooling decisions independently based on local temperature and load information, and achieves collaborative control by sharing global information.

6. The cooling coordinated control method according to claim 5, characterized in that: An adaptive algorithm based on proximal strategy optimization is used to automatically learn the relationship between transformer load changes and temperature changes, including: Construct a state set based on the real-time load change and temperature change of the transformer; According to the adaptive algorithm optimized by the proximal strategy, a cooling resource allocation strategy is constructed for the state set, and it is optimized iteratively; The updated cooling resource allocation strategy is used to calculate the cooling resource allocation amount for each area of ​​the transformer at the next moment.

7. The cooling coordinated control method according to claim 5, characterized in that: Each cooling unit of the transformer makes independent cooling decisions based on local temperature and load information, and achieves collaborative control by sharing global information, including: Each cooling unit of the transformer makes independent cooling decisions based on local temperature and local load information, and independently calculates the cooling resource allocation; Global information sharing among transformer cooling units to build a global state set; The cooling resources of each cooling unit of the transformer are globally optimized through a collaborative control mechanism, and the global cooling resource allocation results are fed back to each cooling unit.

8. The cooling coordinated control method according to claim 5, characterized in that: Each cooling unit of the transformer makes cooling decisions independently based on local temperature and load information, and then achieves collaborative control by sharing global information, including: Monitor the power consumption and cooling efficiency of each cooling unit of the transformer, and dynamically adjust the working status of each cooling unit based on the real-time load of the transformer; When abnormal temperature rise or overheating risk occurs in a specific area of ​​the transformer, the system automatically triggers cooling adjustments through the loop.

9. A cooling cooperative control device, characterized in that: include: A data acquisition module, used to collect and obtain parameter sets; Processing module, used to predict the heat load distribution of each area of ​​the transformer and allocate cooling resources in real time; Energy consumption optimization module, used to monitor the power consumption and cooling efficiency of each cooling unit; Monitoring and feedback module to identify abnormal temperature rise or overheating risk and trigger cooling adjustments.

10. A transformer, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the cooling coordinated control method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the cooling coordinated control method according to any one of claims 1 to 8.

Images