Cabinet heat dissipation method of interlocking system and training method of heat dissipation information determination model

Abstract

The embodiments of this specification provide a cabinet heat dissipation method of an interlocking system and a training method for a heat dissipation information determination model; the cabinet heat dissipation method includes: obtaining the heat dissipation weight coefficient, temperature threshold information, and temperature control priority information of the cage where various control units are located in the target cabinet of the interlocking system, and combining them to obtain heat dissipation management parameters; at any time during the operation of the target cabinet, calling the heat dissipation information determination model to process the heat dissipation management parameters, the ambient temperature information of the target cabinet, the temperature information of multiple locations in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet to obtain auxiliary heat dissipation information of the heat dissipation module; based on the auxiliary heat dissipation information, controlling the heat dissipation module to dissipate heat to the target cabinet. This method can improve the heat dissipation performance of the interlocking system.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of rail transit technology, and in particular to a cabinet heat dissipation method and device for an interlocking system, a training method and device for a heat dissipation information determination model, a computing device, a computer-readable storage medium, and a computer program product. Background Art

[0002] With the development of rail transit technology, the requirements for the operational reliability of various equipment used in rail transit are becoming higher and higher. In rail transit, interlocking systems are used to control railway signals, such as controlling the status of equipment such as switches and signal machines through logical relationships to ensure the safety of train operation.

[0003] The interlocking system includes execution units installed at each station, such as turnout control unit, signal control unit, safety input and output control unit and scattered control units, which are used to control the operation of corresponding equipment. These control units are centrally installed in a cabinet, and different types of control units can be installed in different cages in the cabinet. These control units generate a lot of heat during operation, which may affect the operating stability, reliability and service life of the control units.

[0004] Usually, the cabinet is cooled only by a heat dissipation method that is preset manually. This heat dissipation method has poor flexibility and is difficult to achieve effective heat dissipation of the cabinet in a targeted manner, resulting in poor heat dissipation effect. Summary of the invention

[0005] The embodiments of this specification provide a cabinet heat dissipation method for an interlocking system, which can improve the heat dissipation effect of the cabinet. One or more embodiments of this specification also relate to a cabinet heat dissipation device for an interlocking system, a training method for a heat dissipation information determination model, a training device for a heat dissipation information determination model, a computing device, a computer-readable storage medium, and a computer program product.

[0006] According to one aspect of an embodiment of this specification, a cabinet heat dissipation method of an interlocking system is provided, the method comprising:

[0007] Obtain the heat dissipation weight coefficient, temperature threshold information, and temperature control priority information of the cages where various control units are located in the target cabinet of the interlocking system, and combine them to obtain the heat dissipation management parameters;

[0008] At any time during the operation of the target cabinet, the heat dissipation information determination model is called to process the heat dissipation management parameters, the ambient temperature information of the target cabinet, the temperature information of multiple positions in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet to obtain the auxiliary heat dissipation information of the heat dissipation module; wherein the heat dissipation information determination model is trained based on multiple training samples and training labels, each of the training samples includes a sample heat dissipation management parameter of a sample cabinet, the ambient temperature information, the sample temperature information of each position in the cabinet, and the sample heat dissipation information of the heat dissipation module in the cabinet, and the training label includes the expected heat dissipation information of the heat dissipation module;

[0009] Based on the auxiliary heat dissipation information, the heat dissipation module is controlled to dissipate heat for the target cabinet.

[0010] According to another aspect of an embodiment of this specification, a method for training a heat dissipation information determination model is provided, including:

[0011] Acquire multiple training samples and training labels, wherein each of the training samples includes a sample heat dissipation management parameter of a sample cabinet, ambient temperature information, sample temperature information of each position in the sample cabinet, and sample heat dissipation information of a sample heat dissipation module in the sample cabinet, and the training label includes expected heat dissipation information of the sample heat dissipation module;

[0012] Inputting the plurality of training samples into an initial heat dissipation information determination model to predict reference heat dissipation information of a sample heat dissipation module corresponding to each training sample;

[0013] Based on the comparison result between the reference heat dissipation information and the expected heat dissipation information in the training label corresponding to each training sample, the initial heat dissipation information determination model is adjusted to obtain a trained heat dissipation information determination model.

[0014] According to another aspect of the embodiments of this specification, there is provided a cabinet heat dissipation device of an interlocking system, comprising:

[0015] The first acquisition module is used to obtain the heat dissipation weight coefficient, temperature threshold information and temperature control priority information of the cage where each control unit is located in the target cabinet of the interlocking system, and combine them to obtain the heat dissipation management parameters;

[0016] A processing module, used for calling a heat dissipation information determination model to process the heat dissipation management parameters, the ambient temperature information of the target cabinet, the temperature information of multiple positions in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet at any time during the operation of the target cabinet, so as to obtain auxiliary heat dissipation information of the heat dissipation module; wherein the heat dissipation information determination model is trained based on multiple training samples and training labels, each of the training samples includes a sample heat dissipation management parameter of a sample cabinet, the ambient temperature information, the sample temperature information of each position in the cabinet, and the sample heat dissipation information of the heat dissipation module in the cabinet, and the training label includes the expected heat dissipation information of the heat dissipation module;

[0017] A control module is used to control the heat dissipation module to dissipate heat for the target cabinet based on the auxiliary heat dissipation information.

[0018] According to another aspect of an embodiment of this specification, a training device for a heat dissipation information determination model is provided, comprising:

[0019] an acquisition module, configured to acquire a plurality of training samples and training labels, wherein each of the training samples comprises a sample heat dissipation management parameter of a sample cabinet, ambient temperature information thereof, sample temperature information of each position in the sample cabinet, and sample heat dissipation information of a sample heat dissipation module in the sample cabinet, and the training label comprises expected heat dissipation information of the sample heat dissipation module;

[0020] A prediction module, used to input the plurality of training samples into an initial heat dissipation information determination model, and predict reference heat dissipation information of a sample heat dissipation module corresponding to each training sample;

[0021] The adjustment module is used to adjust the initial heat dissipation information determination model based on the comparison result between the reference heat dissipation information and the expected heat dissipation information in the training label corresponding to each training sample to obtain a trained heat dissipation information determination model.

[0022] According to another aspect of the embodiments of this specification, there is provided a computing device, including: a memory and a processor;

[0023] The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer programs / instructions are executed by the processor, the steps of the above method are implemented.

[0024] According to another aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores a computer program / instruction, and the steps of the above method are implemented when the computer program / instruction is executed by a processor.

[0025] According to another aspect of the embodiments of this specification, a computer program product is provided, including a computer program / instruction, and when the computer program / instruction in the computer program product is executed in a processor, the steps of the above method are implemented.

[0026] In one embodiment of the specification, a heat dissipation information determination model can be trained for a cabinet in an interlocking system, and then the heat dissipation information determination model can be called at any time during the operation of the target cabinet, and the heat dissipation management parameters, the current actual ambient temperature information, the temperature information of multiple positions in the target cabinet, and the current heat dissipation information of the heat dissipation module are processed to obtain auxiliary heat dissipation information adapted to the actual heat situation of the current cabinet for the heat dissipation module. Cooling the target cabinet based on the auxiliary heat dissipation information can achieve a better heat dissipation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the structure of a fully electronic interlocking system;

[0028] Figure 2 It is a structural diagram of the equipment layout of each station in a fully electronic interlocking system;

[0029] Figure 3 It is a schematic diagram of a cabinet of an interlocking execution unit;

[0030] Figure 4 It is a logical relationship diagram between a heat dissipation module, a main control unit and a temperature sensor unit provided in an embodiment of this specification;

[0031] Figure 5 This is a flow chart of a cabinet heat dissipation method of an interlocking system provided in an embodiment of this specification;

[0032] Figure 6 is a flow chart of a method for training a heat dissipation information determination model provided in an embodiment of this specification;

[0033] Figure 7 It is a structural schematic diagram of a cabinet heat dissipation device of an interlocking system provided in an embodiment of this specification;

[0034] Figure 8 It is a structural schematic diagram of a training device for a heat dissipation information determination model provided in an embodiment of this specification;

[0035] Fig. 9 It is a structural block diagram of a computing device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0036] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0037] The terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms of "one", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items. The term "at least one" in one or more embodiments of this specification refers to "one or more", and "multiple" refers to "two or more". The term "including" is an open description and should be understood as "including but not limited to", and may include other content on the basis of the content already described.

[0038] It should be understood that although the terms "first", "second", etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, "first" may also be referred to as "second", and similarly, "second" may also be referred to as "first". Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0039] In the rail transit environment, the interlocking system is an important component used for signal control of rail transit to ensure the safety of train operation. The interlocking system can be a fully electronic interlocking system. Figure 1 This is a schematic diagram of the structure of a fully electronic interlocking system. Figure 1As shown, the fully electronic interlocking system may include an interlocking main control unit (including a series I interlocking logic unit and a series II interlocking logic unit), an I / O execution unit (including a series I and series II turnout control unit, a signal control unit, an input / output control unit, etc.), a control and display computer, and a monitoring computer. Among them, the turnout is used to switch the train from one track to another. The signal is used to provide driving instructions and safety information to the train driver, and indicates whether the train can continue to travel, slow down or stop, etc. through different light colors, shapes or positions. Different types of modules in the I / O execution unit can be located in different electronic terminal cabinets, and these electronic terminal cabinets can be located in a cabinet, corresponding to different cages in the cabinet. The control and display computer may include multiple, such as Figure 1 The control and display computer A and the control and display computer B are used as examples for illustration.

[0040] Each device in the full electronic interlocking system can be connected by a communication network to ensure the continuity and reliability of communication between the devices. For example, the communication network can be a common Ethernet redundant network. Figure 1 As shown, two communication networks can be set up between each device, called network A and network B respectively. For example, the control and display computer is connected to the interlocking logic unit through two control and display networks, and the interlocking logic unit is connected to each I / O execution unit through two control networks. Two maintenance networks are set up between each I / O execution unit of the monitoring computer, and the monitoring computer is also connected to the above two control and display networks to monitor the status of each I / O execution unit and each interlocking logic unit, so as to facilitate timely repair when a component fails or other problems occur.

[0041] Usually, multiple I / O execution units are set up at each station on the rail transit line, and each I / O execution unit can be set up in an I / O execution unit cabinet. Stations with more I / O execution units can be called equipment concentration stations, and stations with fewer I / O execution units can be called non-equipment concentration stations. In each equipment concentration station, in addition to the I / O execution unit, an interlocking master control unit (such as set up in an interlocking master control cabinet) and a monitoring computer can also be arranged. In a non-equipment concentration station, only I / O execution units can be arranged. Figure 2 It is a structural diagram of the equipment layout of each station in a fully electronic interlocking system, and Figure 2 Take the example of each station being a centralized equipment station. Figure 2As shown, each station includes an interlocking main control cabinet and an I / O execution unit cabinet connected in communication, and the interlocking main control cabinets of each station can also communicate with each other. Each interlocking main control cabinet includes a series I interlocking logic unit and a series II interlocking logic unit, which respectively perform corresponding logic control. Each I / O execution unit cabinet can include a switch control unit, a signal control unit, a track circuit and other trackside equipment control unit, and these control units respectively control the operation of the corresponding equipment, such as respectively controlling the operation of the switch, signal, track circuit and other trackside equipment.

[0042] For example, the I / O execution unit in the fully electronic interlocking system (hereinafter referred to as the interlocking execution unit) may include a switch control unit, a signal control unit, a safety input control unit, a safety output control unit and scattered control units, which may be arranged in different cages in a cabinet. Figure 3 It is a schematic diagram of a cabinet of an interlocking execution unit, such as Figure 3 As shown, the cabinet includes cages 301 to 305, which are used to place the above five control units. The traditional gravity relay can be replaced by a board-level safety relay, and the control circuit of the original trackside equipment such as turnouts and signal machines can be designed on a small board. Figure 3 In the example, each vertical bar in the cage corresponds to a board. Due to the need to meet SIL4 safety requirements, each control unit needs to use more board-level safety relays to build the drive acquisition circuit, and a larger power module is required to support the operation of the drive acquisition circuit. Figure 3 In the cabinet, a wiring layer is also provided, which includes multiple power modules, which are used to support the operation of the built drive acquisition circuit. In this way, the heat generated by the entire interlocking execution unit is usually large. If the interlocking execution unit cannot be effectively cooled, it may affect the service life and reliability of each component in the interlocking execution unit, and correspondingly affect the operation of the entire rail transit. Therefore, there is an urgent need for a solution that can stably control the operating environment of the entire execution unit within a relatively ideal range to improve the service life and reliability of the components on the board.

[0043] In one related technology, two sets of interlocking execution units are set up as hot standby for each other. Even if one of the execution units in one set fails, the execution unit in the other set can still be used to work to ensure the safety of the system. In another related technology, a more detailed fault information analysis is performed on the interlocking execution unit so that maintenance personnel can grasp the equipment operation status in time, quickly formulate fault handling measures, and handle the fault in a timely and accurate manner. In both methods, relevant processing methods after the failure of the interlocking execution unit are considered. The failure of the interlocking execution unit cannot be avoided, and it will still have a certain impact on the operation of the interlocking execution unit, and it will cause more equipment and human resources consumption. The heat dissipation effect of the interlocking execution unit still has certain shortcomings.

[0044] The embodiments of this specification provide a cabinet heat dissipation method for an interlocking system, the cabinet is a cabinet for arranging an interlocking execution unit, the method manages the heat dissipation of the cabinet by constructing a scientific heat dissipation model, improves robustness from the perspective of environmental management of thermal management, and realizes thermal management under the requirements of system reliability in the field of industrial control, which can effectively improve the heat dissipation performance of the cabinet and reduce the probability of failure of the interlocking execution unit. The embodiments of this specification also involve a cabinet heat dissipation device for an interlocking system, a training method and device for a heat dissipation information determination model, a computing device, a computer-readable storage medium, and a computer program product, which will be described in detail one by one in the following embodiments.

[0045] The cabinet heat dissipation method of the interlocking system can be applied to the cabinet heat dissipation device and the computing device of the interlocking system. The cabinet heat dissipation device and the computing device of the interlocking system can be devices with information processing functions and communication functions. A heat dissipation module is provided in the cabinet of the interlocking system for dissipating heat from the cabinet. For example, the heat dissipation module can include a fan unit, and the cabinet is cooled by blowing air to various positions in the cabinet. For example, the heat dissipation module can include a plurality of fan units, each of which is used to dissipate heat from a cage. Each fan unit can include 8 fans, which are respectively arranged in two rows in the corresponding cage, and 4 fans are arranged in each row. The fan unit can also be arranged in other ways, or can also include other numbers of fans, which are not limited here. Optionally, the heat dissipation module can be a 1U heat dissipation module, that is, a cooling module designed to be installed in a server chassis with a height of 1U (about 44.45 mm or 1.75 inches).

[0046] The interlocking system may also include a main control unit of the heat dissipation module, which is used to control the heat dissipation module to work, and the main control unit can perform functions such as data acquisition, data output, data calculation and communication. The cabinet heat dissipation method in the embodiment of this specification can be executed by the main control unit. The interlocking system may also include a plurality of temperature sensor units arranged at different positions of the cabinet, which are used to detect the temperature of each position of the cabinet respectively. By way of example, five temperature sensor units are arranged in the cabinet, of which four temperature sensor units are respectively arranged on the edge sides of the cabinet, and the other one is arranged in the central area of ​​the cabinet. The number and arrangement of the temperature sensor units may also be different from this example.

[0047] Figure 4 FIG. 1 is a logical relationship diagram between a heat dissipation module, a main control unit and a temperature sensor unit provided in an embodiment of the present specification. Figure 4 As shown, in the embodiment of the present specification, the temperature sensor unit can transmit the collected temperature information of the monitoring point to the main control unit of the heat dissipation module, and the main control unit can determine the speed information of each fan in the heat dissipation module based on the temperature information, so as to control each fan to run at a corresponding speed based on the speed information, so as to achieve heat dissipation of the cabinet. The main control unit of the heat dissipation module in each cabinet can also communicate with the main control units of other heat dissipation modules, and the communication can mutually transmit heat dissipation-related information to perform corresponding heat dissipation control on the cabinet in combination with the heat dissipation conditions of other cabinets, or the communication can be other information other than heat dissipation.

[0048] Figure 5 This is a flow chart of a cabinet cooling method for an interlocking system provided in an embodiment of this specification. This method can be applied to a cabinet cooling device for an interlocking system, such as the cabinet cooling device being the main control unit of the cooling module in the aforementioned cabinet. Figure 5 As shown, the cabinet heat dissipation method of the interlocking system includes the following steps 502 to 506.

[0049] Step 502: Obtain heat dissipation weight coefficients, temperature threshold information, and temperature control priority information of cages where various control units are located in the target cabinet of the interlocking system, and combine them to obtain heat dissipation management parameters.

[0050] In the embodiment of this specification, the target cabinet can be a cabinet for placing various control units in any station in the interlocking system, and different types of control units are located in different cages of the cabinet. In the embodiment of this specification, the various types of control units include a turnout control unit, a signal control unit, a safety input control unit, a safety output control unit and a scattered control unit as an example, and the cages where these types of control units are located can be respectively called a turnout control cage, a signal control cage, a safety input control cage, a safety output control cage and a scattered control cage. The cabinet heat dissipation device can obtain relevant information about heat dissipation in the cages where various control units are located in the target cabinet, such as these relevant information including heat dissipation weight coefficients, temperature threshold information and temperature control priority information. The cabinet heat dissipation device can determine these relevant information by analyzing the structure of the target cabinet, the specific layout of each component and the characteristics of each component.

[0051] With respect to the heat dissipation weight coefficient, the cabinet heat dissipation device can determine it by analyzing the layout information and heating characteristics of various control units. For example, the cabinet heat dissipation device determines the heat flow data of the cage where each control unit is located based on the layout information of various control units in the target cabinet of the interlocking system and the heating characteristics of various control units; based on the heat flow data, the heat dissipation weight coefficient of the cage where each control unit is located is determined. The layout information of each type of control unit may include the layout information of the board in the corresponding cage, and the layout information of the components on the board.

[0052] The cabinet cooling device can use fluid dynamics simulation tools to perform thermal simulation on each cage based on the layout information and heating characteristics of various control units to obtain the heat flow data of each cage. For example, the cabinet cooling device can analyze the three aspects of heat conduction, convection and radiation for each cage, determine the temperature distribution data of each position due to heat flow, and determine the heat flow data based on the temperature distribution data. Then, for different heat flow data, the initial heat dissipation weight coefficient is configured for each type of cage. For example, the heat flow data may include heat flux density, and the heat dissipation weight coefficient may increase as the heat flux density increases.

[0053] With respect to the temperature threshold information, the cabinet heat dissipation device can determine the temperature threshold information of the cages where the various control units are located based on the target operating environment information of the cages where the various control units are located. The cabinet heat dissipation device can determine the target operating environment information of each type of cage, and the target operating environment information can be pre-set by the staff, such as based on the staff's experience, or can also be derived based on the working characteristics of the control unit in the cage. For example, the target operating environment can be considered to be the optimal operating environment for the cage. Based on the target operating environment information of each type of cage, the temperature threshold information can be set for the cage. , where t min and t max They represent the maximum temperature threshold and the minimum temperature threshold respectively, and the temperature of the cage should be between the two temperature thresholds.

[0054] With respect to the temperature control priority information, the cabinet cooling device can determine the temperature control priority information of the cages where various control units are located based on the impact information when various control units fail. For example, the impact information when various control units fail may include the impact range information or impact degree information when any board in the control unit fails. The impact information may be the impact information for the interlocking system, or it may be the impact information for other equipment. If a failure of a certain control unit affects the signal transmission of multiple stations in the interlocking system, the impact information may include the information of the multiple stations, or it may include the number of the multiple stations. The greater the impact degree and range corresponding to any type of control unit, the higher the temperature control priority can be set for the cage where the control unit is located. .

[0055] The heat dissipation weight coefficient, temperature threshold information, and temperature control priority information of each cage can form a basic single cage heat dissipation model. The cabinet heat dissipation device can combine the heat dissipation weight coefficients, temperature threshold information, and temperature control priority information of the cages where various control units are located to obtain heat dissipation management parameters. For example, the heat dissipation management parameters can be a matrix, such as the initial heat dissipation management operation matrix. The matrix can be , where DC represents the turnout control unit, XH represents the signal control unit, AQSR represents the safety input control unit, AQSC represents the safety output control unit, and LS represents the scattered control unit.

[0056] Step 504: at any time during the operation of the target cabinet, the heat dissipation information determination model is called to process the heat dissipation management parameters, the ambient temperature information of the target cabinet, the temperature information of multiple positions in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet to obtain the auxiliary heat dissipation information of the heat dissipation module.

[0057] Among them, the heat dissipation information determination model is trained based on multiple training samples and training labels. Each training sample includes sample heat dissipation management parameters of a sample cabinet, ambient temperature information, sample temperature information at each position in the cabinet, and sample heat dissipation information of the heat dissipation module in the cabinet. The training label includes the expected heat dissipation information of the heat dissipation module.

[0058] The cabinet heat dissipation device can obtain the current temperature information of multiple positions in the target cabinet at any time during the operation of the target cabinet, which is equivalent to obtaining the temperature information in real time. If the temperature information is the temperature information collected by each temperature sensor unit in the target cabinet, the multiple positions are respectively the monitoring positions of each temperature sensor unit.

[0059] The cabinet cooling device can also obtain the ambient temperature information of the target cabinet at any time, which can be collected by a temperature sensor unit outside the cabinet or input by a staff. The ambient temperature information can be updated in real time or can be a set value.

[0060] The cabinet heat dissipation device can also obtain the heat dissipation information of the heat dissipation module in the target cabinet at any time, and the heat dissipation information can be the operation status information of the heat dissipation module. For example, the heat dissipation module includes fan units corresponding to each cage, and the heat dissipation information can include the rotation speed of each fan in each fan unit. Optionally, the heat dissipation information can also include other information such as the wind direction or size of the fan.

[0061] The cabinet heat dissipation device can call the heat dissipation information determination model for processing based on the latest acquired information to be processed, so as to obtain the auxiliary heat dissipation information of the heat dissipation module. The auxiliary heat dissipation information can indicate the operating status information that the heat dissipation module should have at the next moment, such as the speed information of each fan. Based on the auxiliary heat dissipation information, it can be ensured that the cabinet in the current state is better cooled. In the embodiment of the present specification, the heat dissipation information determination model can be used to determine the auxiliary heat dissipation information of the corresponding fan unit for each cage, and then combine to obtain the auxiliary heat dissipation information of the heat dissipation module as a whole. In some embodiments, the heat dissipation information determination model can also be used to directly determine the auxiliary heat dissipation information of the heat dissipation module as a whole.

[0062] In the embodiment of the present specification, the heat dissipation information determination model may be a neural network model, which includes an input layer, a hidden layer, and an output layer connected in sequence, and the data is input into the input layer of the heat dissipation information determination model, and then processed by the hidden layer and the output layer to obtain auxiliary heat dissipation information. Step 504 may include the following steps s2 to s6.

[0063] Step s2: construct a cage feature vector based on the heat dissipation management parameters, the ambient temperature information of the target cabinet, the temperature information of multiple locations in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet; wherein the heat dissipation information of the heat dissipation module includes the operating status information of the heat dissipation module.

[0064] The cabinet cooling device can first construct a cage feature vector for the acquired information to be processed (i.e., cooling management parameters, ambient temperature information, temperature information of multiple locations in the target cabinet, and cooling information of the cooling module). In the embodiment of this specification, auxiliary cooling information is determined for the fan unit corresponding to the target cage in the cooling module, and the number of temperature sensor units is 5, and the fan unit corresponding to the target cage in the cooling module includes 8 fans. The target cage can be any cage in the cabinet. For example, the cage feature vector ,in, Indicates the real-time temperature data collected by each temperature sensor. Indicates the real-time temperature information of the environment where the target cabinet is located. Indicates the real-time speed data of each fan in the fan unit corresponding to the target cage. Represents the heat dissipation management parameter. The cage feature vector can be used as input data for the heat dissipation information determination model.

[0065] Step s4, inputting the cage feature vector into the input layer of the heat dissipation information determination model, and using the first weight matrix between the input layer and the hidden layer of the heat dissipation information determination model to perform a linear transformation on the cage feature vector, and combining the first activation function to perform a nonlinear transformation on the cage feature vector to obtain output information of the hidden layer; wherein the activation threshold of the first activation function is adjusted based on the first bias vector.

[0066] In the embodiment of the present specification, the input layer of the heat dissipation information determination model may include multiple neural network nodes, such as 15 neural network nodes. The hidden layer includes multiple hidden operation nodes, such as 6 hidden operation nodes. A first weight matrix is ​​constructed between the input layer and the hidden layer. , the first weight matrix can be a 15×6 matrix. The first weight matrix is ​​used to reflect the connection strength between the neurons in the input layer and the neurons in the hidden layer. Each neuron data output by the input layer is at least part of the information in the cage feature vector, and the first weight matrix can be used to perform a linear transformation management of weighted summation on the neuron data. A first activation function is also set for the hidden layer , the cage feature vector can be nonlinearly transformed by using the first activation function. The output information of the hidden layer can be obtained by combining the linear transformation result and the nonlinear transformation result.

[0067] The heat dissipation information determination model also includes a first bias vector , used to process the cage feature vector The activation threshold between each neuron layer (i.e., input layer and hidden layer) is adjusted. The output expression of the hidden layer can be , the output result is It can be output to the next layer of neural network (such as output layer) for operation. The first bias vector is also used to Neuron training is performed to make X satisfy the control of layer-by-layer operation in linear transformation, that is, to ensure that the current operation is related to the input data at the previous moment. Optionally, the hidden layer may include multiple neural network layers, which correspond to the cage feature vectors at multiple moments respectively. The calculation of the cage feature vector at the next moment can be combined with the calculation result of the cage feature vector at the previous moment, and finally the final output information can be output through layer-by-layer iteration. .

[0068] Step s6, using the output layer of the heat dissipation information determination model to process the output information of the hidden layer to obtain auxiliary heat dissipation information of the heat dissipation module output by the output layer; wherein the output layer includes at least one node corresponding to the heat dissipation module.

[0069] For example, the heat dissipation information can be used to determine the second weight matrix and the second bias vector between the hidden layer and the output layer of the model, the output information of the hidden layer can be linearly transformed, and the output information can be nonlinearly transformed in combination with the second activation function to obtain auxiliary heat dissipation information of the heat dissipation module output by the output layer.

[0070] The output information of the hidden layer can be input into the output layer of the heat dissipation information determination model, and the output layer can be constructed based on the structure of the heat dissipation module. For example, the output layer includes multiple control nodes, and the multiple control nodes correspond to each fan in the heat dissipation module, such as the multiple control nodes include 8 fan control nodes. The output layer outputs the auxiliary heat dissipation information of the fan unit corresponding to the target cage in the heat dissipation module. For each fan unit in the heat dissipation module, the corresponding auxiliary heat dissipation information is determined using the above method, and the auxiliary heat dissipation information of the heat dissipation module as a whole can be obtained. If the heat dissipation module only includes one fan unit, the output layer output in step s6 is the auxiliary heat dissipation information of the heat dissipation module as a whole.

[0071] A second weight matrix is ​​set between the hidden layer and the output layer of the heat dissipation information determination model The second weight matrix can be constructed based on the 6 hidden operation nodes in the hidden layer and the 8 fan unit control nodes in the output layer, such as the second weight matrix is ​​a 6×8 matrix. The second weight matrix is ​​used to connect the operation relationship between the neuron operation points in the hidden layer and the fan control nodes in the output layer. The heat dissipation information determination model also has a second bias vector , the second bias vector The parameters of the baseline neurons are trained based on the 8 fan unit control nodes to perform linear transformation operations on the fan unit control nodes. A second activation function is also set for the output layer , the second activation function is used to process the nonlinear transformation operation for the fan control node in the entire operation network. The expression of the output vector of the output layer can be ,After the aforementioned calculations, the auxiliary heat dissipation information can be obtained, such as the target speed of the 8 fan control nodes.

[0072] Step 506: Based on the auxiliary heat dissipation information, control the heat dissipation module to dissipate heat for the target cabinet.

[0073] After obtaining the auxiliary heat dissipation information, the cabinet heat dissipation device can control the heat dissipation module based on the auxiliary heat dissipation information, such as controlling the operation of each fan according to the target speed of each fan in the heat dissipation module. Optionally, the auxiliary heat dissipation information may also include wind direction or other operation information of the fan.

[0074] In one embodiment of the present specification, the cabinet heat dissipation device can perform cabinet heat dissipation control in combination with the cabinet heat dissipation conditions before the current moment. For example, step 506 may include: determining the target heat dissipation information of the heat dissipation module based on the difference information between the auxiliary heat dissipation information and the actual heat dissipation information at the moment before the arbitrary moment; and controlling the heat dissipation module to dissipate heat for the target cabinet based on the target heat dissipation information. The cabinet heat dissipation device can obtain the actual heat dissipation information of the heat dissipation module at each moment during the operation process, and determine the difference information between the auxiliary heat dissipation information and the actual heat dissipation information at that moment, such as the speed difference of the fan. Combined with the difference information at each previous moment, the target heat dissipation information of the heat dissipation module is further determined, and then the heat dissipation module is controlled, so that closed-loop control of the heat dissipation module can be achieved.

[0075] For example, the target heat dissipation information and the difference between the auxiliary heat dissipation information and the actual heat dissipation information of the heat dissipation module at the previous moment can satisfy the set target relationship, and the cabinet heat dissipation device can calculate the corresponding target heat dissipation information based on the target relationship. n Vector PID (Proportion Integration Differentiation) control algorithm, the target relationship includes proportional term, integral term and differential term.

[0076] If the target relationship is target heat dissipation information ,in, Represents the proportional gain parameter, which is used to control the increase or decrease of the fan speed; Represents the difference information between the auxiliary fan speed and the current fan speed in the time dimension (also called error); represents the integral gain parameter, which is used to manage the steady-state error of the accumulated error in the time range [0, k]; represents the cumulative error from the time range [0, k]; It represents the differential gain parameter, which is used to solve the influence of the error change rate on the control output. This gain can be used to make early adjustments for future errors. is the rate of change of the error, that is, the change in error from the previous moment to the current moment.

[0077] Accordingly, the above-mentioned determination of the target heat dissipation information of the heat dissipation module based on the difference information between the auxiliary heat dissipation information at the moment before the arbitrary moment and the actual heat dissipation information may include: adjusting the proportional gain parameter based on the difference information between the auxiliary heat dissipation information at any moment and the actual heat dissipation information; adjusting the integral gain parameter based on the cumulative information of the difference information at each moment before the arbitrary moment; adjusting the differential gain parameter based on the change information of the difference information of the target cabinet at any moment; determining the target heat dissipation information of the heat dissipation module based on the adjusted proportional gain parameter, integral gain parameter and differential gain parameter, as well as the difference information, cumulative information and change information.

[0078] The cabinet cooling device can continuously determine the difference between the auxiliary cooling information of the cooling module and the current actual cooling information, that is, , based on the current runtime , the cumulative information of the difference information at the previous moment, and the change information of the difference information relative to the previous moment, adjust the proportional gain parameter, the integral gain parameter and the differential gain parameter. And based on the adjusted gain parameters, the difference information of the heat dissipation information, the cumulative information of the difference information, the change information of the difference information and the aforementioned target relationship, calculate the target heat dissipation information Based on the target heat dissipation information, closed-loop regulation and control of the heat dissipation situation can be achieved in the time dimension, and the temperature of the entire cabinet can be managed.

[0079] In the embodiments of this specification, a heat dissipation management solution for the execution unit in the full electronic interlocking system is provided. In this solution, the cabinet heat dissipation device can perform closed-loop control on the heat dissipation module according to the real-time monitoring data of the heat dissipation module and the cabinet, and adjust the operating state of the heat dissipation module in real time to optimize the control effect. This solution can solve the problem that the stability and life of the entire execution unit cabinet are affected due to different cabinet configurations, different heat source distribution conditions of the cage, and different heat flow conditions such as heat conduction and convection in the cabinets where the control units are located in different application areas; it can ensure that the appropriate heat dissipation method can be determined for various cabinets to ensure effective heat dissipation of the cabinet.

[0080] In the embodiment of this specification, heat dissipation analysis is performed based on the heat conduction, convection and radiation of different cages in the cabinet, and the initial heat dissipation weight coefficient, temperature threshold information and temperature control priority information are set to construct a basic thermal model inside the cabinet (corresponding to the aforementioned heat dissipation management parameters). Secondly, according to the basic thermal model configuration of different cages, an initial heat dissipation control strategy model (corresponding to the aforementioned cage feature vector) is constructed for the actual heat dissipation of a single cage. After that, the specific heat dissipation strategy for the heat dissipation module is determined through the neural network algorithm and combined with PID intelligent control, thereby realizing the intelligent and dynamic adjustment of the heat dissipation process of the entire cabinet. In the embodiment of this specification, in view of the uneven heat dissipation of different cages in the cabinet, multiple aspects are comprehensively considered, including the distribution of heat sources, the design of heat conduction and convection, intelligent monitoring and control, etc., to construct an intelligent heat dissipation management model. In different application scenarios, a closed-loop control model can be used to complete the intelligent management of the heat dissipation of the entire cabinet, solve the heat dissipation management problem of the full electronic interlocking execution unit, and improve the operating stability, heat dissipation efficiency and equipment service life of the entire full electronic interlocking system.

[0081] In summary, in the cabinet heat dissipation method of the interlocking system provided in the embodiment of this specification, a heat dissipation information determination model can be trained for the cabinet in the interlocking system, and then at any time during the operation of the target cabinet, the heat dissipation information determination model can be called to process the heat dissipation management parameters, the current actual ambient temperature information, the temperature information of multiple positions in the target cabinet, and the current heat dissipation information of the heat dissipation module, and obtain auxiliary heat dissipation information adapted to the actual heat situation of the current cabinet for the heat dissipation module. Cooling the target cabinet based on the auxiliary heat dissipation information can achieve a better heat dissipation effect.

[0082] Before executing the cabinet cooling method of the interlocking system provided in the embodiment of this specification, the cabinet cooling device may first execute the training method of the cooling information determination model to train the cooling information determination model, and then execute the cabinet cooling method using the trained cooling information determination model. Figure 6 A flowchart of a method for training a heat dissipation information determination model provided in an embodiment of the present specification. The training method can be applied to a model training device, which can be the same as the cabinet heat dissipation device for executing the cabinet heat dissipation method described above, or can be different. In different cases, after the heat dissipation information determination model is trained by the model training device, the cabinet heat dissipation device obtains the heat dissipation information determination model to execute the cabinet heat dissipation method described above.

[0083] Figure 6 It is a flowchart of a training method for a heat dissipation information determination model provided in an embodiment of the present specification, which specifically includes the following steps 602 to 606. Figure 6 The steps can refer to the previous Figure 2 The implementation method of the corresponding steps in the above description will not be repeated here.

[0084] Step 602: Obtain multiple training samples and training labels, wherein each training sample includes sample heat dissipation management parameters of the sample cabinet, ambient temperature information, sample temperature information at each position in the sample cabinet, and sample heat dissipation information of a sample heat dissipation module in the sample cabinet, and the training label includes expected heat dissipation information of the sample heat dissipation module.

[0085] The training sample is the data supported by the heat dissipation information determination model for processing, and can be the historical data for which the processing results corresponding to the training task have been obtained. The training label corresponding to the training sample is the correct processing result that should be obtained by executing the training task for the training sample, and is used to determine the prediction loss of the initial heat dissipation information determination model for the training task, thereby adjusting the model parameters of the initial heat dissipation information determination model.

[0086] Step 604: Input multiple training samples into the initial heat dissipation information determination model to predict reference heat dissipation information of the sample heat dissipation module corresponding to each training sample.

[0087] Step 604 is the process of processing the training samples in the model training phase. The specific implementation process is the same as that for Figure 2 Steps s2 to s6 in the introduction are similar, and for details, please refer to the above-mentioned relevant introduction.

[0088] Step 606: Based on the comparison result between the reference heat dissipation information and the expected heat dissipation information in the training label corresponding to each training sample, the initial heat dissipation information determination model is adjusted to obtain a trained heat dissipation information determination model.

[0089] The model training device can adjust the model parameters of the initial heat dissipation information determination model based on the comparison results between the reference heat dissipation information and the expected heat dissipation information in the training label, and return to execute the operation step of obtaining training data (that is, step 602) until the training stop condition is reached, thereby obtaining a trained heat dissipation information determination model.

[0090] Specifically, the loss value can be calculated based on the reference heat dissipation information and the training label, and the model parameters of the initial heat dissipation information determination model can be adjusted by back propagation based on the loss value, and then the execution returns to step 602, and the training samples and training labels are continuously selected to train the initial heat dissipation information determination model until the training stop condition is reached, and the trained heat dissipation information determination model is obtained. The training stop condition can be that the loss value is less than the loss value threshold, or that the number of iterations reaches the number threshold, or that the loss value threshold and the number threshold are combined to determine whether the training stop condition is reached.

[0091] The loss value threshold is the critical value of the loss value. When the loss value is greater than or equal to the loss value threshold, it means that there is still a certain deviation between the prediction result of the initial heat dissipation information determination model and the actual result, and the model parameters of the initial heat dissipation information determination model still need to be adjusted. At this time, it is determined that the training stop condition has not been met; when the loss value is less than the loss value threshold, it means that the prediction result of the initial heat dissipation information determination model is close enough to the actual result, and the training can be stopped. At this time, it can be determined that the training stop condition has been met. The value of the loss value threshold is selected according to the actual situation.

[0092] In addition, the number threshold is set according to the actual situation. When the number of iterations reaches the number threshold, it means that the initial heat dissipation information determination model has been trained enough times. At this time, the prediction result of the initial heat dissipation information determination model is very close to the real result, and the training can be stopped.

[0093] The heat dissipation information determination model obtained by training using the training method provided in the embodiments of this specification can more accurately determine the target heat dissipation information required for effective heat dissipation of the cabinet based on the actual state of the cabinet, thereby ensuring good heat dissipation performance of the cabinet.

[0094] Corresponding to the above-mentioned cabinet heat dissipation method embodiment of the interlocking system, this specification also provides a cabinet heat dissipation device embodiment of the interlocking system, Figure 7 1 is a schematic diagram of the structure of a cabinet heat dissipation device of an interlocking system provided in an embodiment of this specification. Figure 7 As shown, the cabinet heat dissipation device of the interlocking system includes:

[0095] The first acquisition module 702 is used to acquire the heat dissipation weight coefficient, temperature threshold information and temperature control priority information of the cages where various control units are located in the target cabinet of the interlocking system, and combine them to obtain the heat dissipation management parameters;

[0096] The processing module 704 is used to call the heat dissipation information determination model to process the heat dissipation management parameters, the ambient temperature information of the target cabinet, the temperature information of multiple positions in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet at any time during the operation of the target cabinet, so as to obtain the auxiliary heat dissipation information of the heat dissipation module; wherein the heat dissipation information determination model is trained based on multiple training samples and training labels, each training sample includes a sample heat dissipation management parameter of a sample cabinet, the ambient temperature information, the sample temperature information of each position in the cabinet, and the sample heat dissipation information of the heat dissipation module in the cabinet, and the training label includes the expected heat dissipation information of the heat dissipation module;

[0097] The control module 706 is used to control the heat dissipation module to dissipate heat for the target cabinet based on the auxiliary heat dissipation information.

[0098] Optionally, the processing module 704 includes:

[0099] A submodule is constructed to construct a cage feature vector based on heat dissipation management parameters, ambient temperature information of the target cabinet, temperature information of multiple locations in the target cabinet, and heat dissipation information of the heat dissipation module in the target cabinet; wherein the heat dissipation information of the heat dissipation module includes operation status information of the heat dissipation module;

[0100] A first processing submodule is used to input the cage feature vector into an input layer of a heat dissipation information determination model, and use a first weight matrix between the input layer and a hidden layer of the heat dissipation information determination model to perform a linear transformation on the cage feature vector, and perform a nonlinear transformation on the cage feature vector in combination with a first activation function to obtain output information of the hidden layer; wherein an activation threshold of the first activation function is adjusted based on a first bias vector;

[0101] The second processing submodule is used to process the output information of the hidden layer using the output layer of the heat dissipation information determination model to obtain auxiliary heat dissipation information of the heat dissipation module output by the output layer; wherein the output layer includes at least one node corresponding to the heat dissipation module.

[0102] Optionally, the second processing submodule is used for:

[0103] The second weight matrix and the second bias vector between the hidden layer and the output layer of the heat dissipation information determination model are used to perform a linear transformation on the output information of the hidden layer, and the output information is nonlinearly transformed in combination with the second activation function to obtain the auxiliary heat dissipation information of the heat dissipation module output by the output layer.

[0104] Optionally, the control module 706 is used to:

[0105] Determine target heat dissipation information of the heat dissipation module based on difference information between the auxiliary heat dissipation information at a time before the arbitrary time and the actual heat dissipation information;

[0106] Based on the target heat dissipation information, the heat dissipation module is controlled to dissipate heat for the target cabinet.

[0107] Optionally, the control module 706 is used to:

[0108] Adjusting the proportional gain parameter based on the difference information between the auxiliary heat dissipation information and the actual heat dissipation information at any moment;

[0109] Based on the accumulated information of the difference information at each moment before the arbitrary moment, adjusting the integral gain parameter;

[0110] Adjusting the differential gain parameter based on the change information of the difference information of the target cabinet at any time;

[0111] Based on the adjusted proportional gain parameter, integral gain parameter and differential gain parameter, as well as the difference information, accumulation information and change information, target heat dissipation information of the heat dissipation module is determined.

[0112] Optionally, the first acquisition module 702 is used to:

[0113] Based on the layout information of various control units in the target cabinet of the interlocking system and the heating characteristics of various control units, the heat flow data of the cages where various control units are located are determined;

[0114] Based on the heat flow data, determine the heat dissipation weight coefficients of the cages where various control units are located;

[0115] Based on target operating environment information of the cages where the various control units are located, determining temperature threshold information of the cages where the various control units are located;

[0116] Based on the impact information when various control units fail, the temperature control priority information of the cages where the various control units are located is determined.

[0117] Optionally, the cabinet heat dissipation device of the interlocking system further includes:

[0118] A second acquisition module is used to acquire multiple training samples and training labels before calling the heat dissipation information determination model to process the heat dissipation management parameters, the ambient temperature information of the target cabinet, the temperature information of multiple positions in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet, wherein each training sample includes a sample heat dissipation management parameter of the sample cabinet, the ambient temperature information, the sample temperature information of each position in the sample cabinet, and the sample heat dissipation information of the sample heat dissipation module in the sample cabinet, and the training label includes the expected heat dissipation information of the sample heat dissipation module;

[0119] A prediction module, used to input a plurality of training samples into an initial heat dissipation information determination model, and predict reference heat dissipation information of a sample heat dissipation module corresponding to each training sample;

[0120] The adjustment module is used to adjust the initial heat dissipation information determination model based on the comparison result between the reference heat dissipation information and the expected heat dissipation information in the training label corresponding to each training sample, so as to obtain a trained heat dissipation information determination model.

[0121] In summary, in the cabinet heat dissipation device of the interlocking system provided in the embodiment of this specification, a heat dissipation information determination model can be trained for the cabinet in the interlocking system, and then at any time during the operation of the target cabinet, the heat dissipation information determination model can be called to process the heat dissipation management parameters, the current actual ambient temperature information, the temperature information of multiple positions in the target cabinet, and the current heat dissipation information of the heat dissipation module, and obtain auxiliary heat dissipation information adapted to the actual heat situation of the current cabinet for the heat dissipation module. Cooling the target cabinet based on the auxiliary heat dissipation information can achieve a better heat dissipation effect.

[0122] Corresponding to the above method embodiment, this specification also provides a training device embodiment for determining a heat dissipation information model. Figure 8 FIG. 1 is a schematic diagram of a training device for a heat dissipation information determination model provided in an embodiment of the present specification. Figure 8 As shown, the device comprises:

[0123] An acquisition module 802 is used to acquire multiple training samples and training labels, wherein each training sample includes a sample heat dissipation management parameter of a sample cabinet, ambient temperature information, sample temperature information of each position in the sample cabinet, and sample heat dissipation information of a sample heat dissipation module in the sample cabinet, and the training label includes expected heat dissipation information of the sample heat dissipation module;

[0124] A prediction module 804 is used to input a plurality of training samples into an initial heat dissipation information determination model to predict reference heat dissipation information of a sample heat dissipation module corresponding to each training sample;

[0125] The adjustment module 806 is used to adjust the initial heat dissipation information determination model based on the comparison result between the reference heat dissipation information and the expected heat dissipation information in the training label corresponding to each training sample, so as to obtain a trained heat dissipation information determination model.

[0126] The above is a schematic scheme of a training device for a heat dissipation information determination model of this embodiment. It should be noted that the technical scheme of the training device for heat dissipation information determination model and the technical scheme of the training method for heat dissipation information determination model belong to the same concept, and the details not described in detail in the technical scheme of the training device for heat dissipation information determination model can be found in the description of the technical scheme of the training method for heat dissipation information determination model.

[0127] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. As for the display device of the target image, since it is basically similar to the embodiment of the display method of the target image, the description is relatively simple, and the relevant parts can be referred to the partial description of the embodiment of the display method of the target image.

[0128] Fig. 9 9 is a block diagram of a computing device provided in an embodiment of the present specification. The components of the computing device 900 include but are not limited to a memory 910 and a processor 920. The processor 920 is connected to the memory 910 via a bus 930, and a database 950 is used to store data.

[0129] The computing device 900 also includes an access device 940 that enables the computing device 900 to communicate via one or more networks 960. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 940 may include one or more of any type of network interface (e.g., a network interface card (NIC)) that is wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a world-wide interoperability for microwave access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, and a near field communication (NFC).

[0130] In one embodiment of the present specification, the above components of the computing device 900 and Fig. 9 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Fig. 9 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0131] The computing device 900 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 900 may also be a mobile or stationary server.

[0132] The processor 920 is used to execute computer programs / instructions, which, when executed by the processor, implement the above Figure 5 or Figure 6 The method shown.

[0133] As for the computing device embodiment, since it is basically similar to the above-mentioned method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0134] One embodiment of the present specification also provides a computer-readable storage medium, which stores computer instructions, and when the computer instructions are executed by the processor, the steps of the above-mentioned image processing method are implemented. The computer instructions include computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable storage medium may include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media do not include electric carrier signals and telecommunication signals.

[0135] An embodiment of the present specification also provides a computer program product, including a computer program / instruction, which implements the steps of the above method when the computer program / instruction is executed in a processor.

[0136] As for the computer-readable storage medium embodiment and the computer program product embodiment, since they are basically similar to the above-mentioned method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0137] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0138] It should be noted that the above is a description of a specific embodiment of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of the present specification.

[0139] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0140] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can understand and use this specification well.

Claims

1. A cabinet heat dissipation method for an interlocking system, characterized in that: The method comprises: Obtaining heat dissipation weight coefficients, temperature threshold information, and temperature control priority information of the cages where various control units are located in the target cabinet of the interlocking system, and combining them to obtain heat dissipation management parameters; wherein the heat dissipation weight coefficients are determined based on the heat flow data of the cages where the various control units are located, and the temperature control priority information is determined based on the impact information when the various control units fail; At any time during the operation of the target cabinet, the heat dissipation information determination model is called to process the heat dissipation management parameters, the ambient temperature information of the target cabinet, the temperature information of multiple positions in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet to obtain the auxiliary heat dissipation information of the heat dissipation module; wherein the heat dissipation information determination model is trained based on multiple training samples and training labels, each of the training samples includes a sample heat dissipation management parameter of a sample cabinet, the ambient temperature information, the sample temperature information of each position in the cabinet, and the sample heat dissipation information of the heat dissipation module in the cabinet, and the training label includes the expected heat dissipation information of the heat dissipation module; Based on the auxiliary heat dissipation information, the heat dissipation module is controlled to dissipate heat for the target cabinet.

2. The method according to claim 1, characterized in that The calling of the heat dissipation information determination model processes the heat dissipation management parameter, the ambient temperature information of the target cabinet, the temperature information of multiple positions in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet to obtain the auxiliary heat dissipation information of the heat dissipation module, including: Constructing a cage feature vector based on the heat dissipation management parameter, the ambient temperature information of the target cabinet, the temperature information of multiple locations in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet; wherein the heat dissipation information of the heat dissipation module includes the operating status information of the heat dissipation module; Inputting the cage feature vector into an input layer of a heat dissipation information determination model, and using a first weight matrix between the input layer and a hidden layer of the heat dissipation information determination model to perform a linear transformation on the cage feature vector, and performing a nonlinear transformation on the cage feature vector in combination with a first activation function to obtain output information of the hidden layer; wherein an activation threshold of the first activation function is adjusted based on a first bias vector; The output information of the hidden layer is processed using the output layer of the heat dissipation information determination model to obtain auxiliary heat dissipation information of the heat dissipation module output by the output layer; wherein the output layer includes at least one node corresponding to the heat dissipation module.

3. The method according to claim 2, characterized in that The step of using the output layer of the heat dissipation information determination model to process the output information of the hidden layer to obtain the auxiliary heat dissipation information of the heat dissipation module output by the output layer includes: The output information of the hidden layer is linearly transformed by utilizing the second weight matrix and the second bias vector between the hidden layer and the output layer of the heat dissipation information determination model, and the output information is nonlinearly transformed by combining with the second activation function to obtain the auxiliary heat dissipation information of the heat dissipation module output by the output layer.

4. The method according to any one of claims 1 to 3, characterized in that: The controlling the heat dissipation module to dissipate heat for the target cabinet based on the auxiliary heat dissipation information includes: Determining target heat dissipation information of the heat dissipation module based on difference information between the auxiliary heat dissipation information at a time before the any time and the actual heat dissipation information; Based on the target heat dissipation information, the heat dissipation module is controlled to dissipate heat for the target cabinet.

5. The method according to claim 4, characterized in that The determining the target heat dissipation information of the heat dissipation module based on the difference information between the auxiliary heat dissipation information at the time before the any time and the actual heat dissipation information includes: Adjusting a proportional gain parameter based on difference information between the auxiliary heat dissipation information and the actual heat dissipation information at any time; Adjusting an integral gain parameter based on the accumulated information of the difference information at each time before the arbitrary time; Adjusting a differential gain parameter based on change information of the difference information of the target cabinet at the any time; Based on the adjusted proportional gain parameter, the integral gain parameter, and the differential gain parameter, as well as the difference information, the accumulation information, and the change information, target heat dissipation information of the heat dissipation module is determined.

6. The method according to any one of claims 1 to 3, characterized in that: The step of obtaining the heat dissipation weight coefficient, temperature threshold information, and temperature control priority information of the cages where various control units are located in the target cabinet of the interlocking system includes: Based on the layout information of various control units in the target cabinet of the interlocking system and the heating characteristics of various control units, the heat flow data of the cages where various control units are located are determined; Based on the heat flow data, determining the heat dissipation weight coefficients of the cages where the various control units are located; Based on the target operating environment information of the cages where the various control units are located, determining the temperature threshold information of the cages where the various control units are located; Based on the impact information when the various control units fail, the temperature control priority information of the cages where the various control units are located is determined.

7. The method according to any one of claims 1 to 3, characterized in that: Before calling the heat dissipation information determination model to process the heat dissipation management parameters, the ambient temperature information of the target cabinet, the temperature information of multiple locations in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet, the method further includes: Acquire multiple training samples and training labels, wherein each of the training samples includes sample heat dissipation management parameters of the sample cabinet, ambient temperature information, sample temperature information of each position in the sample cabinet, and sample heat dissipation information of a sample heat dissipation module in the sample cabinet, and the training label includes expected heat dissipation information of the sample heat dissipation module; Inputting the plurality of training samples into an initial heat dissipation information determination model to predict reference heat dissipation information of a sample heat dissipation module corresponding to each training sample; Based on the comparison result between the reference heat dissipation information and the expected heat dissipation information in the training label corresponding to each training sample, the initial heat dissipation information determination model is adjusted to obtain a trained heat dissipation information determination model.

8. A method for training a heat dissipation information determination model, characterized in that: The heat dissipation information determination model is the heat dissipation information determination model described in any one of claims 1 to 7, and the training method comprises: Acquire multiple training samples and training labels, wherein each of the training samples includes a sample heat dissipation management parameter of a sample cabinet, ambient temperature information, sample temperature information of each position in the sample cabinet, and sample heat dissipation information of a sample heat dissipation module in the sample cabinet, and the training label includes expected heat dissipation information of the sample heat dissipation module; Inputting the plurality of training samples into an initial heat dissipation information determination model to predict reference heat dissipation information of a sample heat dissipation module corresponding to each training sample; Based on the comparison result between the reference heat dissipation information and the expected heat dissipation information in the training label corresponding to each training sample, the initial heat dissipation information determination model is adjusted to obtain a trained heat dissipation information determination model.

9. A cabinet heat dissipation device for an interlocking system, characterized in that: include: The first acquisition module is used to obtain the heat dissipation weight coefficient, temperature threshold information and temperature control priority information of the cage where each type of control unit is located in the target cabinet of the interlocking system, and combine them to obtain the heat dissipation management parameters; wherein the heat dissipation weight coefficient is determined based on the heat flow data of the cage where the various types of control units are located, and the temperature control priority information is determined based on the impact information when the various types of control units fail; A processing module, used for calling a heat dissipation information determination model to process the heat dissipation management parameters, the ambient temperature information of the target cabinet, the temperature information of multiple positions in the target cabinet, and the heat dissipation information of the heat dissipation module in the target cabinet at any time during the operation of the target cabinet, so as to obtain auxiliary heat dissipation information of the heat dissipation module; wherein the heat dissipation information determination model is trained based on multiple training samples and training labels, each of the training samples includes a sample heat dissipation management parameter of a sample cabinet, the ambient temperature information, the sample temperature information of each position in the cabinet, and the sample heat dissipation information of the heat dissipation module in the cabinet, and the training label includes the expected heat dissipation information of the heat dissipation module; A control module is used to control the heat dissipation module to dissipate heat for the target cabinet based on the auxiliary heat dissipation information.

10. A training device for a heat dissipation information determination model, characterized in that: The heat dissipation information determination model is the heat dissipation information determination model described in any one of claims 1 to 7, and the training device comprises: an acquisition module, configured to acquire a plurality of training samples and training labels, wherein each of the training samples comprises a sample heat dissipation management parameter of a sample cabinet, ambient temperature information thereof, sample temperature information of each position in the sample cabinet, and sample heat dissipation information of a sample heat dissipation module in the sample cabinet, and the training label comprises expected heat dissipation information of the sample heat dissipation module; A prediction module, used to input the plurality of training samples into an initial heat dissipation information determination model, and predict reference heat dissipation information of a sample heat dissipation module corresponding to each training sample; The adjustment module is used to adjust the initial heat dissipation information determination model based on the comparison result between the reference heat dissipation information and the expected heat dissipation information in the training label corresponding to each training sample to obtain a trained heat dissipation information determination model.

11. A computing device, characterized in that: include: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer program / instructions are executed by the processor, the method according to any one of claims 1 to 8 is implemented.

12. A computer-readable storage medium, characterized in that: A computer program / instruction is stored, and when the computer program / instruction is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

13. A computer program product, characterized in that The method comprises a computer program / instruction, which implements the method according to any one of claims 1 to 8 when executed by a processor.

Images