Temperature control method and device of memory, storage equipment and storage medium

Abstract

The invention provides a temperature control method and device of a memory, storage equipment and a storage medium, and relates to the technical field of memorizers, and the method comprises the following steps: obtaining real-time temperature data of the memory; inputting the real-time temperature data into a preset adaptive prediction model, performing reinforcement learning on the real-time temperature data based on the adaptive prediction model, and outputting decision information; and executing temperature control operation on the memory according to the decision information. By adopting the memory temperature control method and device, the temperature control effect on the memory can be improved, and adaptive temperature control on the memory in a complex temperature environment is realized.

Description

Technical Field

[0001] The present application relates to the field of memory technology, and in particular to a temperature control method, device, storage device and storage medium for a memory. Background Art

[0002] In the related art, the temperature control of the memory is usually carried out by using the proportional, integral and differential (PID) industrial control method. However, since the PID algorithm is relatively dependent on empirical parameters and requires manual adjustment of relevant parameters, when the temperature of the memory in a complex temperature environment is controlled, the PID algorithm that relies on empirical parameters is very complex and is very easy to over-adjust when controlling the temperature rise and fall of the memory, resulting in poor temperature control effect of the memory. Summary of the invention

[0003] The main purpose of the embodiments of the present application is to propose a temperature control method, apparatus, storage device, medium and product for a memory, aiming to improve the effect of temperature control on the memory and realize adaptive temperature control of the memory in a complex temperature environment.

[0004] To achieve the above object, a first aspect of an embodiment of the present application provides a temperature control method for a memory, the method comprising:

[0005] Get real-time temperature data of the storage;

[0006] Inputting the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and output decision information;

[0007] A temperature control operation on the memory is performed according to the decision information.

[0008] In some embodiments, the decision information includes: a temperature difference between predicted temperature data of the memory and target operating temperature data of the memory; wherein the predicted temperature data is temperature data predicted after a temperature control operation is performed on the memory to increase or decrease the temperature of the memory;

[0009] The performing a temperature control operation on the memory according to the decision information includes:

[0010] Comparing the temperature difference with a preset temperature difference limit value to obtain a comparison result; wherein the temperature difference limit value is determined based on a temperature control interval of the memory;

[0011] A temperature control operation is performed on the memory based on the comparison result, so that the temperature data of the memory after the temperature is increased / decreased matches the target operating temperature data.

[0012] In some embodiments, before inputting the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and outputting decision information, the method further includes:

[0013] Acquiring temperature data of the memory after performing temperature control operations on the memory multiple times to increase / decrease the temperature of the memory;

[0014] Perform reinforcement learning model training based on the plurality of temperature data and a preset temperature control cycle to obtain the adaptive prediction model;

[0015] The temperature control period is a control period used when a temperature control operation is performed on the memory.

[0016] In some embodiments, the acquiring of the temperature data of the memory after performing temperature control operations on the memory multiple times to increase / decrease the temperature of the memory includes at least one of the following:

[0017] Acquire multiple temperature data of the memory from a preset local data storage space; wherein the temperature sensor of the memory collects temperature data of the memory after each heating / cooling operation in response to the temperature control operation, and stores the collected temperature data in the local data storage space;

[0018] A plurality of temperature data of the memory is obtained from a preset server platform; wherein the server platform records the temperature data of the memory after each heating / cooling operation in response to the temperature control operation.

[0019] In some embodiments, the reinforcement learning model training is performed based on the plurality of temperature data and a preset temperature control cycle to obtain the adaptive prediction model, including:

[0020] Calculating the instantaneous temperature change rate of the memory based on two adjacent temperature data among the plurality of temperature data and the temperature control period;

[0021] Calculating a temperature change acceleration of the memory based on the instantaneous temperature change rate and the temperature control period;

[0022] Reinforcement learning model training is performed based on the temperature change acceleration, the instantaneous temperature change rate and the temperature control period to obtain the adaptive prediction model.

[0023] In some embodiments, the inputting the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and output decision information includes:

[0024] Comparing the real-time temperature data with the target operating temperature data of the memory to obtain a comparison result;

[0025] When the comparison result indicates that the real-time temperature data does not match the target operating temperature data, the real-time temperature data is input into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and output decision information.

[0026] In some embodiments, after comparing the real-time temperature data with the target operating temperature data of the memory to obtain a comparison result, the method further includes:

[0027] When the comparison result indicates that the real-time temperature data matches the target operating temperature data, a preset equilibrium state control is performed on the memory so that the temperature data of the memory is maintained at the target operating temperature data, or the temperature data of the memory is maintained within an allowable fluctuation range near the target operating temperature data.

[0028] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a temperature control device for a memory, the device comprising:

[0029] An acquisition module is used to acquire real-time temperature data of the memory;

[0030] An adaptive model prediction module, used for inputting the real-time temperature data into a preset adaptive prediction model, so as to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model, and output decision information;

[0031] A temperature control module is used to perform a temperature control operation on the memory according to the decision information.

[0032] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes a storage device, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the method provided in the first aspect above when executing the computer program.

[0033] To achieve the above objectives, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the method provided in the first aspect above.

[0034] To achieve the above objectives, the fifth aspect of the embodiments of the present application proposes a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the method provided in the first aspect above.

[0035] The temperature control method of the memory, the temperature control device of the memory, the storage device, the computer-readable storage medium and the computer program product proposed in the embodiments of the present application obtain real-time temperature data of the memory; input the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model, and output decision information; and perform temperature control operations on the memory according to the decision information.

[0036] Thus, compared with the traditional PID control method for temperature control of the memory, the embodiment of the present application only needs to obtain the real-time temperature data of the memory, and then perform reinforcement learning on the real-time temperature data based on the pre-built adaptive prediction model, so as to obtain decision information on how to perform temperature control on the memory, and then directly perform temperature control operations on the memory according to the decision information. Thus, the embodiment of the present application can improve the effect of temperature control on the memory, and can realize adaptive temperature control of the memory in a complex temperature environment based only on the temperature data of the memory.

[0037] In addition, the embodiment of the present application generates a decision on temperature control of the memory based on an adaptive prediction model, and does not require manual parameter adjustment like the traditional PID algorithm, thereby greatly optimizing the computational complexity of temperature control of the memory, shortening the decision time on how to temperature control the memory, and thus achieving shorter and more flexible temperature control of the memory. Experiments conducted in actual complex temperature environments show that the use of the embodiment of the present application can effectively avoid the problem of temperature overshoot that is prone to occur when using traditional PID control to increase or decrease the temperature of the memory, and the use of the embodiment of the present application can enter the equilibrium state faster than using the traditional PID control method to control the temperature of the memory. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A schematic flow chart of steps in some embodiments of the temperature control method of a memory provided in an embodiment of the present application;

[0039] Figure 2 A schematic diagram of a temperature control system architecture involved in some embodiments of the temperature control method for a memory provided in an embodiment of the present application;

[0040] Figure 3 for Figure 1Schematic diagram of detailed process flow of step S103;

[0041] Figure 4 A schematic flow chart of steps for constructing an adaptive prediction model in some embodiments of the temperature control method for a memory provided in an embodiment of the present application;

[0042] Figure 5 for Figure 4 Schematic diagram of detailed process flow of step S402;

[0043] Figure 6 for Figure 1 Schematic diagram of detailed process flow of step S102;

[0044] Figure 7 A schematic diagram of the structural framework of a temperature control device for a memory provided in an embodiment of the present application in some embodiments;

[0045] Figure 8 A schematic diagram of the internal structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0047] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0049] Before introducing the embodiments of the present application, the overall concept of the embodiments of the present application is first described here.

[0050] With the rapid development of electronic equipment technology, the application environment of electronic equipment is becoming more and more extensive. Based on this, how to control the temperature of the memory used in electronic equipment in a complex temperature environment to ensure the performance of the memory is the key to ensure the efficient and stable operation of electronic equipment.

[0051] In the related art, the temperature control of the memory is usually carried out by the PID control method. However, since the PID algorithm is relatively dependent on empirical parameters and requires manual adjustment of relevant parameters, when the temperature of the memory in a complex temperature environment is controlled, the PID algorithm that relies on empirical parameters is very complex and is very easy to over-adjust when controlling the temperature rise and fall of the memory, resulting in poor temperature control effect of the memory.

[0052] To this end, an embodiment of the present application obtains real-time temperature data of a memory; inputs the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model, and outputs decision information; and performs temperature control operations on the memory according to the decision information.

[0053] Thus, compared with the traditional PID control method for temperature control of the memory, the embodiment of the present application only needs to obtain the real-time temperature data of the memory, and then perform reinforcement learning on the real-time temperature data based on the pre-built adaptive prediction model, so as to obtain decision information on how to perform temperature control on the memory, and then directly perform temperature control operations on the memory according to the decision information. Thus, the embodiment of the present application can improve the effect of temperature control on the memory, and can realize adaptive temperature control of the memory in a complex temperature environment based only on the temperature data of the memory.

[0054] It should be understood that the temperature control method of the memory provided in the embodiment of the present application can be applied to the terminal, can also be applied to the server side, and can also be software running in the terminal or the server side. In some embodiments, the terminal can be a smart phone, tablet computer, laptop computer, desktop computer, etc. equipped with a memory; the server side can be an independent physical server equipped with a memory, or a server cluster or distributed system composed of multiple physical servers equipped with a memory, or a cloud server equipped with a memory that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms and other basic cloud computing services; the software can be an application that implements the temperature control method of the memory, etc., but is not limited to the above forms.

[0055] Alternatively, the present application may also be used in numerous general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present application may be described in the general context of computer executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types.

[0056] For ease of understanding and explanation, the following text will take the terminal device applying the temperature control method of the memory provided by the embodiment of the present application as an example to explain the present application in detail. The implementation of the temperature control method of the memory provided by the embodiment of the present application in any of the above-mentioned forms of the subject application can refer to the process of the terminal device applying the steam pressure to control the temperature control method of the memory described below.

[0057] Please refer to Figure 1 , Figure 1 The temperature control method of the memory provided in the present application is a flowchart of the steps in some embodiments. It should be understood that although Figure 1 The execution order of some method steps is shown in the figure, but based on different design requirements of actual applications, the temperature control method of the memory provided in the embodiment of the present application can certainly adopt an execution order different from the method steps shown in the figure. That is, Figure 1 The order of the steps of the method shown does not constitute a limitation on the execution logic order of the temperature control method of the memory provided in the embodiment of the present application. Figure 1 Reasonable changes in the sequence of steps of the method shown should be included in the protection scope of the temperature control method of the memory provided in the embodiment of the present application.

[0058] like Figure 1 As shown, in some embodiments, the temperature control method of the memory provided in the embodiments of the present application may include steps S101 to S103.

[0059] Step S101: Acquire real-time temperature data of the memory.

[0060] When the terminal device performs temperature control on the memory in a complex temperature scenario, it first collects real-time temperature data of the memory based on the temperature sensor of the memory.

[0061] It should be noted that a complex temperature scenario can be an application scenario where the temperature of the environment in which the memory is located changes greatly in a short period of time. For example, for a memory configured in an electronic device installed near a vehicle engine, the temperature of the environment in which the memory is located is relatively low before the vehicle is started, but after the vehicle is started, the temperature of the environment in which the memory is located will rise rapidly in a short period of time. This is a complex temperature change scenario for the memory.

[0062] In addition, the real-time temperature data may include the temperature value of the memory acquired in real time by the temperature sensor, and the time information when the temperature sensor acquires the temperature value. For ease of understanding and explanation, the temperature data mentioned below only include the actual temperature value.

[0063] In some embodiments, the terminal device may also collect real-time temperature data of the memory through a temperature sensor of the memory after performing a temperature control operation on the memory each time to increase / decrease the temperature of the memory.

[0064] It should be noted that if Figure 2 As shown, in the embodiment of the present application, the terminal device can specifically adopt the following Figure 2 The system architecture shown performs temperature control on the memory. That is, the terminal device directly collects the real-time temperature data of the memory through the temperature sensor, and then transmits the real-time temperature data to the controller, and the controller issues a control instruction to the memory based on the real-time temperature data to increase / cool the memory. Moreover, in the process of the controller controlling the temperature increase / cooling of the memory, the temperature sensor still monitors the temperature data of the memory in real time, and transmits the temperature data to the controller, and the controller continues to perform the next step of temperature control operation on the memory based on the temperature data.

[0065] Step S102: inputting the real-time temperature data into a preset adaptive prediction model, so as to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model, and output decision information.

[0066] It should be noted that the terminal device can construct an adaptive prediction model based on the reinforcement learning algorithm in advance, and then store the adaptive prediction model locally for any time. Alternatively, the terminal device can also initiate a request to the server platform through the communication connection with the server platform of the storage so that the server platform returns the adaptive prediction model to the terminal device according to the device type and other identifiers of the storage actually controlled by the terminal device. Among them, the server platform can construct an adaptive prediction model based on the reinforcement learning algorithm in advance on the local side, and then store the adaptive prediction model locally for each terminal device to call.

[0067] After the terminal device obtains the real-time temperature data of the memory, it immediately calls the pre-built adaptive prediction model, and then inputs the real-time temperature data into the adaptive prediction model, thereby using the real-time temperature data for reinforcement learning based on the adaptive prediction model, and then based on the result of reinforcement learning on the real-time temperature data, outputs decision information indicating how to control the temperature of the memory.

[0068] In some embodiments, the terminal device uses an adaptive prediction model to perform reinforcement learning on the real-time temperature data of the memory, and can predict the distance data between the temperature data of the memory after the temperature is increased / or decreased and the optimal operating temperature data of the memory based on the real-time temperature data if the temperature control operation is currently performed on the memory. Then, the adaptive prediction model can directly output the distance data as decision information, or the adaptive prediction model can further determine whether it is necessary to control the temperature of the memory to increase or to control the temperature of the memory based on the distance data, and output the determined result as decision information.

[0069] It should be noted that the optimal operating temperature data of the memory may be the operating temperature when the memory exhibits the optimal operating performance. For example, the optimal operating temperature data of the memory may be any temperature value between 40° and 45°. It should be understood that based on different design requirements of actual applications, in different feasible implementations, the optimal operating temperature data of the memory may of course be different due to differences in type and working environment. The temperature control method of the memory provided in the embodiment of the present application does not limit the specific size of the optimal operating temperature data of the memory.

[0070] Step S103: performing a temperature control operation on the memory according to the decision information.

[0071] After the terminal device performs reinforcement learning on the real-time temperature data of the memory based on the adaptive prediction model to obtain the decision information output by the adaptive prediction model, it immediately performs temperature control operations on the memory according to the decision information to increase / cool the storage.

[0072] It should be noted that the terminal device can perform a temperature control operation on the memory to increase the temperature of the memory when the real-time temperature data of the memory is lower than the optimal operating temperature data of the memory, and can perform a temperature control operation on the memory to reduce the temperature of the memory when the real-time temperature data of the memory is higher than the optimal operating temperature data of the memory.

[0073] In some embodiments, when the decision information output by the adaptive prediction model is the distance data between the temperature data of the memory and the optimal working temperature data of the memory, the terminal device can send instructions to the memory to control the memory to heat up / cool down with the maximum power when the distance data indicates that the temperature data of the memory after heating / cooling is far away from the optimal working temperature data of the memory. Alternatively, when the distance data indicates that the temperature data of the memory after heating / cooling is close to the optimal working temperature data of the memory, the terminal device can send instructions to the memory to control the memory to heat up / cool down with the minimum power (or control the memory to stop heating / cooling and automatically heat up / cool down only relying on the residual power of the previous heating / cooling).

[0074] In other embodiments, when the decision information is to control the temperature of the memory, the terminal device directly sends instructions to the memory according to the decision information to control the temperature of the memory to increase. Alternatively, when the decision information is to control the temperature of the memory to decrease, the terminal device directly sends instructions to the memory according to the decision information to control the temperature of the memory to decrease.

[0075] In an embodiment of the present application, when temperature control is performed on a memory in a complex temperature scenario through a terminal device, first, real-time temperature data of the memory is collected based on the temperature sensor of the memory, and then a pre-built adaptive prediction model is called, and then the real-time temperature data is input into the adaptive prediction model, so that the real-time temperature data is used for reinforcement learning based on the adaptive prediction model, and then based on the result of reinforcement learning on the real-time temperature data, decision information indicating how to control the temperature of the memory is output, so that the temperature control operation of the memory is performed according to the decision information, so that the storage is heated up / cooled down.

[0076] Thus, compared with the traditional PID control method for temperature control of the memory, the embodiment of the present application only needs to obtain the real-time temperature data of the memory, and then perform reinforcement learning on the real-time temperature data based on the pre-built adaptive prediction model, so as to obtain decision information on how to perform temperature control on the memory, and then directly perform temperature control operations on the memory according to the decision information. Thus, the embodiment of the present application can improve the effect of temperature control on the memory, and can realize adaptive temperature control of the memory in a complex temperature environment based only on the temperature data of the memory.

[0077] In addition, since the embodiment of the present application generates a decision on temperature control of the memory based on an adaptive prediction model, there is no need to manually adjust parameters like the traditional PID algorithm, which can greatly optimize the computational complexity of temperature control of the memory, shorten the decision time on how to control the temperature of the memory, and thus achieve a shorter cycle and more flexible temperature control of the memory. Combined with the experimental results conducted in an actual complex temperature environment, the embodiment of the present application can effectively avoid the problem of temperature overshoot that is easy to occur when using traditional PID control to increase and decrease the temperature of the memory, and the embodiment of the present application can enter the equilibrium state faster than the traditional PID control method to control the temperature of the memory.

[0078] In some embodiments, the decision information includes the temperature difference between the predicted temperature data of the memory and the target operating temperature data of the memory (the above-mentioned distance data). The predicted temperature data is the temperature data predicted by the terminal device after the temperature control operation is performed on the memory to increase / decrease the temperature of the memory, that is, the terminal device calls the adaptive prediction model based on the real-time temperature data of the memory to predict the current temperature data of the memory after the temperature control operation is performed on the memory. In addition, the target operating temperature data of the memory is the optimal operating temperature data of the above-mentioned memory.

[0079] Please refer to Figure 3 , Figure 3 for Figure 1 Schematic diagram of the detailed process flow of step S103.

[0080] like Figure 3 As shown, in some embodiments, the above step S103: performing the temperature control operation on the memory according to the decision information may include steps S301 and S302 as shown below.

[0081] Step S301: Compare the temperature difference value with a preset temperature difference limit value to obtain a comparison result; wherein the temperature difference limit value is determined based on a temperature control range of the memory.

[0082] It should be noted that the temperature control interval of the memory can be an acceptable temperature control interval for the memory to support the electronic device to implement actual engineering in a complex temperature environment. The temperature difference limit value includes an upper limit value and a lower limit value. The terminal device can use the extreme value of the temperature control interval as the temperature difference limit value. For example, assuming that the temperature control interval is (lower threshold, upper threshold), the terminal device can directly use the lower threshold as the lower limit value of the temperature difference limit value, and use the upper threshold as the upper limit value of the temperature difference limit value.

[0083] After obtaining the temperature difference between the predicted temperature data of the memory output by the adaptive prediction model and the target operating temperature data of the memory, the terminal device immediately compares the temperature difference with the upper limit value of the temperature difference boundary value to obtain a comparison result, and compares the temperature difference with the lower limit value of the temperature difference boundary value to obtain a comparison result.

[0084] Step S302: performing a temperature control operation on the memory based on the comparison result, so that the temperature data of the memory after heating / cooling matches the target operating temperature data.

[0085] It should be noted that the matching of the temperature data of the memory with the target operating temperature data may be that the temperature data of the memory is equal to the target operating temperature data, or that the difference between the temperature data of the memory and the target operating temperature data is within a preset allowable fluctuation range. The specific value of the allowable fluctuation range may be determined based on the accuracy of the temperature sensor of the memory in measuring the temperature of the memory. For example, the allowable fluctuation range may specifically be (-0.5°, 0.5°). In addition, by making the difference between the real-time temperature data of the memory and the target operating temperature data within a preset allowable fluctuation range, it can be determined that the temperature control of the memory has entered a balanced state, that is, the memory is able to perform optimal performance for data access in a temperature environment close to the target operating temperature data.

[0086] After the terminal device obtains the comparison result, if the temperature difference is less than the lower limit value of the temperature difference limit value, the terminal device executes temperature control operation 1: sends an instruction to the memory to control the heating of the memory, so that after the memory is heated up, the temperature data of the memory collected by the temperature sensor in real time matches the target operating temperature data.

[0087] In addition, after the terminal device obtains the comparison result, if the temperature difference is greater than the upper limit value of the temperature difference limit value, the terminal device executes temperature control operation 2: sends instructions to the memory to control the memory to dissipate heat, so that after the memory is dissipated and cooled, the temperature data of the memory collected in real time by the temperature sensor matches the target operating temperature data.

[0088] Furthermore, after obtaining the comparison result, if the temperature difference is less than the upper limit of the temperature difference limit but greater than the lower limit of the temperature difference limit, the terminal device executes temperature control operation 3: sending an instruction to the memory to control the memory to stop further heating / cooling.

[0089] For example, after obtaining the temperature difference, the terminal device may refer to the control logic shown below to determine the temperature control operation to be performed on the memory, specifically which of the above-mentioned temperature control operations 1, 2 or 3.

[0090]

[0091] Among them, action is the temperature control operation currently executed on the memory determined by the terminal device, enableheating is to control the memory heating to increase the temperature, disable cooling is to control the memory heat dissipation to cool down, and decision is the temperature difference.

[0092] In some embodiments, considering that the heating (temperature increase) power and the heat dissipation (temperature cooling) power of the memory may not be completely equal, the terminal device can determine a buffer area where the memory temperature changes relatively slowly by controlling the memory to turn on heating and heat dissipation at the same time.

[0093] In other embodiments, the terminal device may also ignore the impact of the difference between the heating power and the heat dissipation power of the memory, and thus does not set a buffer area to directly set the upper and lower limits of the temperature difference limit value to "0".

[0094] In this embodiment, the terminal device compares the temperature difference between the predicted temperature data of the memory output by the adaptive prediction model and the target operating temperature data of the memory with the upper and lower limits of the temperature difference limit value to obtain a corresponding comparison result, and then determines whether to control the memory to increase the temperature or to control the memory to decrease the temperature based on the size relationship between the temperature difference indicated by the comparison result and the upper and lower limits. In this way, compared with the traditional method, this embodiment only needs to perform the temperature control operation on the memory based on the adaptive prediction model through the terminal device, so that there is no need to manually adjust the parameters, which greatly optimizes the calculation complexity of the temperature control of the memory, shortens the decision time of how to control the temperature of the memory, and thus can achieve a shorter cycle and more flexible temperature control of the memory.

[0095] Please refer to Figure 4 , Figure 4 A schematic flowchart of the steps of constructing an adaptive prediction model involved in some embodiments of the temperature control method of the memory provided in the embodiments of the present application.

[0096] like Figure 4As shown, in some embodiments, the temperature control method of the memory provided in the embodiment of the present application may further include steps S401 and S402 as shown below before the above-mentioned step S102: inputting the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and outputting decision information.

[0097] Step S401: Acquire temperature data of the memory after performing temperature control operations on the memory multiple times to increase / decrease the temperature of the memory.

[0098] When building an adaptive prediction model, the terminal device first obtains multiple temperature data of the memory as model training data. Among them, each temperature data of the memory is the temperature value collected by the temperature sensor of the memory after the memory is heated up / cooled by performing a temperature control operation on the memory. By repeatedly performing the temperature control operation on the memory to heat up / cool the memory, and collecting temperature data each time after the memory is heated up / cooled by the temperature sensor of the memory, multiple temperature data of the memory can be obtained.

[0099] In some embodiments, the terminal device may store the temperature data of the memory locally, and then directly read the temperature data as model training data when building the adaptive prediction model. Alternatively, when building the adaptive prediction model, the terminal device may also obtain the temperature data of the memory from the server platform of the memory as model training data. Based on this, when executing the above step S401, the terminal device may execute at least one of the following methods 1 and 2.

[0100] Method 1: Acquire multiple temperature data of the memory from a preset local data storage space; wherein, the temperature sensor of the memory collects the temperature data of the memory after each heating / cooling operation in response to the temperature control operation, and stores the collected temperature data in the local data storage space.

[0101] It should be noted that the local data storage space can be a data storage space pre-configured by the terminal device for recording and storing data generated during the self-running process. For example, the terminal device can divide a part of the data storage space in the above-mentioned memory as the local data storage interval. For another example, the terminal device can also specially configure another storage medium independent of the above-mentioned memory as the local data storage space.

[0102] The terminal device performs temperature control operations on the memory multiple times, thereby controlling the heating of the memory to increase the temperature or controlling the heat dissipation of the memory to decrease the temperature. Moreover, each time the terminal device performs temperature control operations on the memory, the temperature data of the memory collected by the temperature sensor after the memory is heated up / cooled based on the temperature control operation is recorded in the local data storage space. In this way, the terminal device can obtain multiple temperature data of the memory from the local data storage space when constructing the adaptive prediction model.

[0103] Method 2: Acquire multiple temperature data of the memory from a preset server platform; wherein the server platform records the temperature data of the memory after each heating / cooling operation in response to the temperature control operation.

[0104] It should be noted that the server platform may be a server platform of the storage device. The server platform may be connected to a terminal device equipped with the storage device by communication, thereby providing various background services related to the storage device to the terminal device based on the communication connection.

[0105] In some embodiments, the terminal device can also communicate with the server platform of the storage device, and based on sending a request to the server platform to obtain multiple temperature data of the storage device, the server platform returns multiple temperature data of the storage device of the same type or model to the terminal device in response to the request. Among them, before releasing the storage device to the outside, the server platform can perform experiments on the storage device in advance to perform temperature control operations on the storage device multiple times to increase or decrease the temperature of the storage device, and each time the storage device is controlled to increase or decrease the temperature, the temperature data of the storage device after the temperature increase / decrease is recorded locally.

[0106] Step S402: Perform reinforcement learning model training based on the multiple temperature data and the preset temperature control cycle to obtain the adaptive prediction model.

[0107] It should be noted that the preset temperature control cycle is the control cycle used when performing temperature control operations on the memory. When the terminal device performs temperature control operations on the memory multiple times to control the temperature increase or decrease of the memory, the control cycle used when each temperature control operation is performed on the memory is synchronously recorded. Usually, the control cycles used by the terminal device to perform temperature control operations on the memory multiple times are the same.

[0108] After acquiring multiple temperature data from the memory, the terminal device uses a reinforcement learning algorithm to perform reinforcement learning model training on the multiple temperature data and temperature control cycles, thereby constructing an adaptive prediction model.

[0109] Please refer to Figure 5 , Figure 5 for Figure 4 Schematic diagram of the detailed step flow of step S402.

[0110] like Figure 5 As shown, in some embodiments, the above-mentioned step S402: performing reinforcement learning model training based on the multiple temperature data and the preset temperature control cycle to obtain the adaptive prediction model may include steps S501 to S503 as described below.

[0111] Step S501: Calculate the instantaneous temperature change rate of the memory based on two adjacent temperature data among the plurality of temperature data and the temperature control cycle.

[0112] When the terminal device performs reinforcement learning model training based on multiple temperature data and temperature control cycles, it first calculates multiple instantaneous temperature change rates of the memory based on the multiple temperature data and temperature control cycles. That is, the terminal device extracts two adjacent temperature data from the multiple temperature data each time, and then performs calculation based on the two adjacent temperature data and the temperature control cycle, thereby obtaining multiple instantaneous temperature change rates that are continuous in time.

[0113] Exemplarily, the terminal device may use the calculation formula shown below to repeat the calculation multiple times, thereby obtaining multiple instantaneous temperature change rates of the memory.

[0114]

[0115] Among them, vtemp is the instantaneous temperature change rate of the memory calculated by the terminal device each time. CurrtentTemperature is the temperature data with a later collection time among the two adjacent temperature data extracted by the terminal device each time, and LastRecordedTemperature is the temperature data with an earlier collection time among the two adjacent temperature data extracted by the terminal device each time. For example, when the terminal device calculates the instantaneous temperature change rate based on the real-time temperature data of the current memory and the temperature data of the memory after the previous control of the memory heating / cooling, CurrtentTemperature is the real-time temperature data of the current memory, and LastRecordedTemperature is the temperature data of the memory after the previous control of the memory heating / cooling. Δt is the control cycle used when the terminal device performs a temperature control operation on the memory to heat up / cool down the memory.

[0116] Step S502: Calculate the temperature change acceleration of the memory based on the instantaneous temperature change rate and the temperature control period.

[0117] After calculating multiple continuous instantaneous temperature change rates of the memory, the terminal device extracts two adjacent instantaneous temperature change rates and performs further calculations based on the two adjacent instantaneous temperature change rates and the temperature control cycle to obtain the temperature change acceleration of the memory during the process of controlling the memory to heat up / cool down.

[0118] Exemplarily, the terminal device may use the formula shown below to further calculate the temperature change acceleration of the memory.

[0119]

[0120] Among them, atemp is the temperature change acceleration of the memory, vtemp1 is the instantaneous temperature change rate with a later time between two adjacent instantaneous temperature change rates, and vtemp2 is the instantaneous temperature change rate with a earlier time between two adjacent instantaneous temperature change rates.

[0121] Step S503: Perform reinforcement learning model training based on the temperature change acceleration, the instantaneous temperature change rate and the temperature control period to obtain the adaptive prediction model.

[0122] After calculating the instantaneous temperature change rate and temperature change acceleration of the memory, the terminal device can construct a reinforcement learning model based on the temperature change acceleration, the instantaneous temperature change rate and the temperature control period and perform reinforcement learning model training, thereby ultimately obtaining an adaptive prediction model based on the real-time temperature data of the memory to adaptively predict how to perform temperature control operations on the memory (such as controlling the memory to heat up / controlling the memory to cool down), and output corresponding decision information.

[0123] Exemplarily, the reinforcement learning model constructed by the terminal device is as follows: vtempΔt+1 / 2atemp(Δt) 2 Based on this, the terminal device can further use the target operating temperature and real-time temperature data of the memory to construct an adaptive prediction model: decision = TargetTemperature-(CurrentTemperature+vtempΔt+1 / 2atemp(Δt) 2 ). Wherein, decision is the decision information (such as the temperature difference between the target operating temperature of the memory and the predicted temperature) output by the adaptive prediction model based on the real-time temperature data of the memory, TargetTemperature is the target operating temperature of the memory, and CurrentTemperature is the real-time temperature data of the memory. The adaptive prediction model firstly calculates the temperature of the memory based on the real-time temperature data of the memory and the model vtempΔt+1 / 2atemp(Δt)2 Calculate the predicted temperature of the memory: CurrentTemperature+vtempΔt+1 / 2atemp(Δt) 2 Then, the adaptive prediction model can obtain decision information decision based on the target operating temperature TargetTemperature of the memory minus the predicted temperature, and the decision information decision represents the temperature difference between the predicted temperature and the target operating temperature of the memory after Δt time.

[0124] It should be noted that the terminal device can continuously calculate the instantaneous temperature change rate and temperature change acceleration of the memory during multiple times of controlling the memory to heat up / cool down, and then combine the instantaneous temperature change rate, temperature change acceleration, temperature control cycle, and the real-time temperature data and target operating temperature data of the memory to construct the above-mentioned adaptive prediction model.

[0125] In this embodiment, the terminal device performs temperature control operations on the memory multiple times, thereby obtaining multiple temperature data of the memory. Afterwards, a reinforcement learning algorithm is used to train the multiple temperature data and the temperature control cycle to perform reinforcement learning model training, thereby constructing an adaptive prediction model. In this way, when the temperature of the memory is subsequently controlled, the terminal device only needs to obtain the real-time temperature data of the memory, and then perform reinforcement learning processing on the real-time temperature data based on the pre-built adaptive prediction model, so as to obtain decision information on how to perform temperature control on the memory, so that the temperature control operation on the memory can be directly performed according to the decision information. In this way, the decision to control the temperature of the memory is generated based on the adaptive prediction model, and there is no need to manually adjust the parameters like the traditional PID algorithm, thereby greatly optimizing the computational complexity of the temperature control of the memory, shortening the decision time of how to control the temperature of the memory, and then realizing a shorter and more flexible temperature control of the memory. Combined with the experimental results conducted in the actual complex temperature environment, the use of the adaptive prediction model to generate the decision to control the temperature of the memory can effectively avoid the problem of temperature overshoot that is easy to occur when the traditional PID control is used to control the temperature of the memory, and can also enable the temperature control of the memory to enter the equilibrium state faster.

[0126] In some embodiments, after obtaining the real-time temperature data of the memory each time, the terminal device may also determine in advance whether the current temperature control of the memory has entered a balanced state based on the real-time temperature data. If it is determined that the temperature control of the memory has not entered a balanced state, the above steps S102 and S103 are executed to further control the temperature increase / cooling of the memory. On the contrary, if it is determined that the temperature control of the memory has entered a balanced state, the real-time temperature data of the memory is maintained.

[0127] Based on this, please refer to Figure 6 , Figure 6 for Figure 1 Schematic diagram of the detailed step flow of step S102.

[0128] like Figure 6 As shown, in some embodiments, the above-mentioned step S102: inputting the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and output decision information may include steps S601 and S602 as shown below.

[0129] Step S601: Compare the real-time temperature data with the target operating temperature data of the memory to obtain a comparison result.

[0130] When the terminal device performs temperature control on the memory, after acquiring the real-time temperature data of the memory through the temperature sensor of the memory, the real-time temperature data is first compared with the target operating temperature data of the memory to obtain a comparison result. The comparison result between the real-time temperature data and the target operating temperature data can at least represent the temperature difference between the real-time temperature data and the target operating temperature data.

[0131] Step S602: When the comparison result indicates that the real-time temperature data does not match the target operating temperature data, the real-time temperature data is input into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and output decision information.

[0132] After the terminal device obtains the comparison result between the real-time temperature data and the target operating temperature data, if the comparison result indicates that the real-time temperature data does not match the target operating temperature data, the terminal device determines that the current temperature control of the memory has not entered an equilibrium state. In this case, the terminal device inputs the real-time temperature data into the adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adapted prediction model, outputs decision information, and performs the subsequent process of temperature control of the memory as embodied by the above-mentioned step S103 and its detailed step flow.

[0133] It should be noted that if the comparison result between the real-time temperature data and the target operating temperature data is that the temperature difference is not equal to 0, or the temperature difference is greater than the upper limit value of the temperature difference limit value, or the temperature difference is less than the lower limit value of the temperature difference limit value, then it indicates that the real-time temperature data does not match the target operating temperature data.

[0134] In some embodiments, the temperature control method of the memory provided in the embodiments of the present application further includes the following steps after the above step S601:

[0135] When the comparison result indicates that the real-time temperature data matches the target operating temperature data, a preset equilibrium state control is performed on the memory so that the temperature data of the memory is maintained at the target operating temperature data, or the temperature data of the memory is maintained within an allowable fluctuation range near the target operating temperature data.

[0136] After the terminal device obtains the comparison result between the real-time temperature data and the target operating temperature data, if the comparison result indicates that the real-time temperature data matches the target operating temperature data, the terminal device determines that the current temperature control of the memory has entered an equilibrium state, so that the terminal device no longer continues to execute steps S102 and S103 to perform temperature control operations on the memory, but instead performs a preset equilibrium state control on the memory, so that the temperature data of the memory can always be kept consistent with the target operating temperature. Even if it is impossible to keep the temperature data of the memory consistent with the target operating temperature, at least the temperature data of the memory is always kept within the allowable fluctuation range near the target operating temperature data.

[0137] It should be noted that if the comparison result between the real-time temperature data and the target operating temperature data is that the temperature difference is equal to 0, or the temperature difference is less than the upper limit value of the temperature difference limit value and greater than the lower limit value of the temperature difference limit value, it indicates that the real-time temperature data matches the target operating temperature data.

[0138] In addition, the preset equilibrium state control can be that the terminal device does not continue to control the memory to heat up / cool down, so that the memory only maintains its real-time temperature data unchanged or fluctuates within a very small range based on the residual potential of the previous heating / cooling. Moreover, the equilibrium state control can also be that when the real-time temperature data of the memory of the terminal device continues to change (affected by the residual potential of the previous heating / cooling), once the temperature difference between the real-time temperature data and the target operating temperature data is greater than the upper limit value of the temperature difference limit value, the terminal device controls the memory to dissipate heat with a relatively small power to cool down, so that the real-time temperature data remains within the allowable fluctuation range; or, once the temperature difference between the real-time temperature data and the target operating temperature data is less than the lower limit value of the temperature difference limit value, the terminal device controls the memory to heat with a relatively small power to heat up, so that the real-time temperature data remains within the allowable fluctuation range.

[0139] In this embodiment, when the terminal device controls the temperature rise / cooling of the memory, once it is found that the real-time temperature data of the memory matches the target operating temperature data, the memory is controlled in a balanced state, so that the temperature data of the memory can always be kept consistent with the target operating temperature. Even if it is not possible to keep the temperature data of the memory consistent with the target operating temperature, at least the temperature data of the memory is always kept within the allowable fluctuation range near the target operating temperature data. In this way, the problem of temperature overshoot when controlling the temperature rise / cooling of the memory can be further avoided, so that the memory can perform data storage, writing and reading processing in the target operating temperature environment, thereby ensuring that the memory can perform optimal performance.

[0140] Based on the same inventive concept, the embodiment of the present application also provides a memory temperature control device for implementing the memory temperature control method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of the temperature control device for one or more memories provided below can refer to the limitations of the memory temperature control method above, and will not be repeated here.

[0141] In one embodiment, Figure 7 As shown, a temperature control device for a memory is provided, comprising: an acquisition module 701, an adaptive model prediction module 702 and a temperature control module 703, wherein:

[0142] An acquisition module 701 is used to acquire real-time temperature data of the memory;

[0143] An adaptive model prediction module 702, used to input the real-time temperature data into a preset adaptive prediction model, to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model, and to output decision information;

[0144] The temperature control module 703 is used to perform a temperature control operation on the memory according to the decision information.

[0145] In some embodiments, the decision information includes: a temperature difference between predicted temperature data of the memory and target operating temperature data of the memory; wherein the predicted temperature data is temperature data predicted after a temperature control operation is performed on the memory to increase or decrease the temperature of the memory;

[0146] The temperature control module 703 is also used to compare the temperature difference with a preset temperature difference limit value to obtain a comparison result; wherein the temperature difference limit value is determined based on the temperature control range of the memory; and based on the comparison result, a temperature control operation on the memory is performed so that the temperature data of the memory after heating / cooling matches the target operating temperature data.

[0147] In some embodiments, the temperature control device of the memory provided by the embodiment of the present application further includes:

[0148] A model building module is used to obtain the temperature data of the memory after performing temperature control operations on the memory multiple times to increase / cool the memory; and to perform reinforcement learning model training based on the multiple temperature data and a preset temperature control cycle to obtain the adaptive prediction model; wherein the temperature control cycle is the control cycle used when performing temperature control operations on the memory.

[0149] In some embodiments, the model building module is also used to obtain multiple temperature data of the memory from a preset local data storage space; wherein the temperature sensor of the memory collects temperature data of the memory after each heating / cooling operation in response to a temperature control operation, and stores the collected temperature data in the local data storage space; and, obtain multiple temperature data of the memory from a preset server platform; wherein the server platform records the temperature data of the memory after each heating / cooling operation in response to a temperature control operation.

[0150] In some embodiments, the model building module is also used to calculate the instantaneous temperature change rate of the memory based on two adjacent temperature data among the multiple temperature data and the temperature control cycle; calculate the temperature change acceleration of the memory based on the instantaneous temperature change rate and the temperature control cycle; and perform reinforcement learning model training based on the temperature change acceleration, the instantaneous temperature change rate and the temperature control cycle to obtain the adaptive prediction model.

[0151] In some embodiments, the adaptive model prediction module 702 is also used to compare the real-time temperature data with the target operating temperature data of the memory to obtain a comparison result; and, when the comparison result indicates that the real-time temperature data does not match the target operating temperature data, the real-time temperature data is input into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and output decision information.

[0152] In some embodiments, the temperature control module 703 is also used to perform a preset equilibrium state control on the memory when the comparison result indicates that the real-time temperature data matches the target operating temperature data, so that the temperature data of the memory remains at the target operating temperature data, or, to keep the temperature data of the memory within an allowable fluctuation range near the target operating temperature data.

[0153] Each module in the temperature control device of the memory provided in the embodiment of the present application can be implemented in whole or in part by software, hardware, or a combination thereof. The above modules can be embedded in or independent of the processor in the electronic device in the form of hardware, or can be stored in the memory in the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0154] In some embodiments, the present application provides an electronic device, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown. The electronic device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the electronic device is used to store model parameters related to the liquid metal reactor. The input / output interface of the electronic device is used to exchange information between the processor and an external device. The communication interface of the electronic device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a temperature control method of a memory is implemented.

[0155] Those skilled in the art will understand that Figure 8The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0156] In some embodiments, an electronic device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0157] Get real-time temperature data of the storage;

[0158] Inputting the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and output decision information;

[0159] A temperature control operation on the memory is performed according to the decision information.

[0160] In some embodiments, the decision information includes: a temperature difference between predicted temperature data of the memory and target operating temperature data of the memory; wherein the predicted temperature data is temperature data predicted after a temperature control operation is performed on the memory to increase or decrease the temperature of the memory;

[0161] When the processor executes the computer program, the following steps are also implemented:

[0162] Comparing the temperature difference with a preset temperature difference limit value to obtain a comparison result; wherein the temperature difference limit value is determined based on a temperature control interval of the memory;

[0163] A temperature control operation is performed on the memory based on the comparison result, so that the temperature data of the memory after the temperature is increased / decreased matches the target operating temperature data.

[0164] In some embodiments, the processor further implements the following steps when executing the computer program:

[0165] Acquiring temperature data of the memory after performing temperature control operations on the memory multiple times to increase / decrease the temperature of the memory;

[0166] Perform reinforcement learning model training based on the plurality of temperature data and a preset temperature control cycle to obtain the adaptive prediction model;

[0167] The temperature control period is a control period used when a temperature control operation is performed on the memory.

[0168] In some embodiments, the processor further implements at least one of the following steps when executing the computer program:

[0169] Acquire multiple temperature data of the memory from a preset local data storage space; wherein the temperature sensor of the memory collects temperature data of the memory after each heating / cooling operation in response to the temperature control operation, and stores the collected temperature data in the local data storage space;

[0170] A plurality of temperature data of the memory is obtained from a preset server platform; wherein the server platform records the temperature data of the memory after each heating / cooling operation in response to the temperature control operation.

[0171] In some embodiments, the processor further implements the following steps when executing the computer program:

[0172] Calculating the instantaneous temperature change rate of the memory based on two adjacent temperature data among the plurality of temperature data and the temperature control period;

[0173] Calculating a temperature change acceleration of the memory based on the instantaneous temperature change rate and the temperature control period;

[0174] Reinforcement learning model training is performed based on the temperature change acceleration, the instantaneous temperature change rate and the temperature control period to obtain the adaptive prediction model.

[0175] In some embodiments, the processor further implements the following steps when executing the computer program:

[0176] Comparing the real-time temperature data with the target operating temperature data of the memory to obtain a comparison result;

[0177] When the comparison result indicates that the real-time temperature data does not match the target operating temperature data, the real-time temperature data is input into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and output decision information.

[0178] In some embodiments, the processor further implements the following steps when executing the computer program:

[0179] When the comparison result indicates that the real-time temperature data matches the target operating temperature data, a preset equilibrium state control is performed on the memory so that the temperature data of the memory is maintained at the target operating temperature data, or the temperature data of the memory is maintained within an allowable fluctuation range near the target operating temperature data.

[0180] In some embodiments, a storage device is provided, which includes a memory and a processor. A computer program is stored in the memory. The steps implemented when the computer program is executed by the processor are the same as the steps implemented when the processor in the above-mentioned electronic device executes the computer program. The same content will not be repeated here.

[0181] In some embodiments, a computer-readable storage medium is provided on which a computer program is stored. The steps implemented when the computer program is executed by a processor are the same as the steps implemented when the processor in the above-mentioned electronic device executes the computer program, and the same content will not be repeated here.

[0182] In one embodiment, a computer program product is provided, including a computer program. The steps implemented when the computer program is executed by a processor are the same as the steps implemented when the processor in the above-mentioned electronic device executes the computer program, and the same contents will not be repeated here.

[0183] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0184] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0185] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A temperature control method for a memory, characterized in that: The method comprises: Get real-time temperature data of the storage; Inputting the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and output decision information; A temperature control operation on the memory is performed according to the decision information.

2. The method according to claim 1, characterized in that The decision information includes: a temperature difference between the predicted temperature data of the memory and the target operating temperature data of the memory; wherein the predicted temperature data is the temperature data predicted after the temperature control operation is performed on the memory to increase / decrease the temperature of the memory; The performing a temperature control operation on the memory according to the decision information includes: Comparing the temperature difference with a preset temperature difference limit value to obtain a comparison result; wherein the temperature difference limit value is determined based on a temperature control interval of the memory; A temperature control operation is performed on the memory based on the comparison result, so that the temperature data of the memory after the temperature is increased / decreased matches the target operating temperature data.

3. The method according to claim 1, characterized in that Before inputting the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and outputting decision information, the method further includes: Acquiring temperature data of the memory after performing temperature control operations on the memory multiple times to increase / decrease the temperature of the memory; Perform reinforcement learning model training based on the plurality of temperature data and a preset temperature control cycle to obtain the adaptive prediction model; The temperature control period is a control period used when a temperature control operation is performed on the memory.

4. The method according to claim 3, characterized in that The acquiring of the temperature data of the memory after performing temperature control operations on the memory multiple times to increase / decrease the temperature of the memory includes at least one of the following: Acquire multiple temperature data of the memory from a preset local data storage space; wherein the temperature sensor of the memory collects temperature data of the memory after each heating / cooling operation in response to the temperature control operation, and stores the collected temperature data in the local data storage space; A plurality of temperature data of the memory is obtained from a preset server platform; wherein the server platform records the temperature data of the memory after each heating / cooling operation in response to the temperature control operation.

5. The method according to claim 3, characterized in that: The step of performing reinforcement learning model training based on the plurality of temperature data and a preset temperature control cycle to obtain the adaptive prediction model includes: Calculating the instantaneous temperature change rate of the memory based on two adjacent temperature data among the plurality of temperature data and the temperature control period; Calculating the temperature change acceleration of the memory based on the instantaneous temperature change rate and the temperature control period; Reinforcement learning model training is performed based on the temperature change acceleration, the instantaneous temperature change rate and the temperature control period to obtain the adaptive prediction model.

6. The method according to claim 1, characterized in that The step of inputting the real-time temperature data into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and outputting decision information includes: Comparing the real-time temperature data with the target operating temperature data of the memory to obtain a comparison result; When the comparison result indicates that the real-time temperature data does not match the target operating temperature data, the real-time temperature data is input into a preset adaptive prediction model to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model and output decision information.

7. The method according to claim 6, characterized in that After the real-time temperature data is compared with the target operating temperature data of the memory to obtain a comparison result, the method further includes: When the comparison result indicates that the real-time temperature data matches the target operating temperature data, a preset equilibrium state control is performed on the memory so that the temperature data of the memory is maintained at the target operating temperature data, or the temperature data of the memory is maintained within an allowable fluctuation range near the target operating temperature data.

8. A temperature control device for a storage device, characterized in that: The device comprises: An acquisition module is used to acquire real-time temperature data of the memory; An adaptive model prediction module, used for inputting the real-time temperature data into a preset adaptive prediction model, so as to perform reinforcement learning on the real-time temperature data based on the adaptive prediction model, and output decision information; A temperature control module is used to perform a temperature control operation on the memory according to the decision information.

9. A storage device, characterized in that: The storage device comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images