Distributed photovoltaic computer computing power control method and system

Abstract

The invention relates to a distributed photovoltaic computer computing power control method and system, and the method comprises the steps: obtaining photovoltaic power generation output power data and computer CPU (Central Processing Unit) computing power demand data, and carrying out the preprocessing; establishing a correlation model of photovoltaic power generation output power data and computer CPU computing power demand; constructing a function of the computer CPU utilization rate and the computer CPU power consumption; according to the correlation model of the photovoltaic power generation output power data and the computer CPU computing power demand, the computer CPU utilization rate and the computer CPU power consumption function, the computer CPU computing power is regulated and controlled; and optimizing the correlation model of the photovoltaic power generation output power data and the computer CPU computing power demand according to the photovoltaic power generation output power data, the computer CPU power consumption and the computer CPU resource use condition in the operation process. The invention provides a distributed photovoltaic computing power control method and system, which can utilize new energy to the maximum extent and reduce the energy consumption of a computing system.

Description

Background Art

[0002] With the transformation of the global energy structure and the promotion of sustainable development goals, as well as the growth of energy demand and the emphasis on clean energy, photovoltaic power generation has broad application prospects, which can reduce dependence on traditional fossil fuels and reduce greenhouse gas emissions. Distributed photovoltaics have also been widely developed as a photovoltaic power generation method. Distributed photovoltaic systems can generate electricity near the user end, reducing transmission losses and other problems. However, distributed photovoltaic power generation devices are volatile and unstable, which can easily have a significant impact on the stability and scheduling difficulty of the power grid.

[0003] In the prior art, battery energy storage systems are usually used for relay storage of electrical energy, thereby reducing the volatility of distributed photovoltaic power generation devices. In addition, different types of photovoltaic modules exhibit different performance characteristics under different lighting and temperature conditions. By combining multiple photovoltaic modules, the entire photovoltaic power generation system can maintain a relatively stable output under a wider range of environmental conditions.

[0004] However, in the above technical solution, the use of a battery energy storage system for relay storage of electrical energy requires the construction and maintenance of a battery energy storage station, thereby increasing the cost of power generation; and the use of different types of photovoltaic modules requires special design according to different environments, thereby increasing the construction cost and design difficulty of the photovoltaic power generation system.

[0005] Therefore, it is necessary to improve one or more problems existing in the above-mentioned related technical solutions.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0007] The purpose of the embodiments of the present disclosure is to provide a computer computing power control method and system for distributed photovoltaics, thereby overcoming one or more problems caused by limitations and defects of related technologies at least to a certain extent.

[0008] In a first aspect, the present application provides a distributed photovoltaic computer computing power control method, comprising:

[0009] Acquire photovoltaic power output data and computer CPU computing power demand data, and perform preprocessing; the computer CPU computing power demand data includes computer CPU usage rate, task queue length and memory usage;

[0010] Establishing a correlation model between photovoltaic power generation output power data and computer CPU computing power demand data according to the processed photovoltaic power generation output power data and the computer CPU computing power demand data;

[0011] Constructing a function of the computer CPU usage rate and the computer CPU power consumption;

[0012] According to the photovoltaic power generation output power data and the computer CPU computing power demand association model and the computer CPU usage rate and the computer CPU power consumption function, the computer CPU computing power is regulated;

[0013] According to the photovoltaic power generation output power data, computer CPU computing power demand data and computer CPU power consumption during operation, the photovoltaic power generation output power data and computer CPU computing power demand association model and the computer CPU utilization rate and computer CPU power consumption function are optimized.

[0014] In a possible implementation manner, the step of obtaining photovoltaic power generation output power data and computer CPU computing power demand data and performing preprocessing includes:

[0015] The output voltage and output current are collected by sensors arranged in the photovoltaic modules to obtain the photovoltaic power generation output power data;

[0016] Obtain computer CPU computing power demand data through the computer task manager;

[0017] The photovoltaic power generation output power data and the computer CPU computing power demand data are preprocessed to obtain a processed data set.

[0018] In a possible implementation manner, the step of preprocessing the photovoltaic power generation output power data and the computer CPU computing power demand data to obtain a processed data set includes:

[0019] Processing missing values ​​and abnormal values ​​of the photovoltaic power generation output power data and the computer CPU computing power demand data;

[0020] The photovoltaic power generation output power data is merged with the computer CPU computing power demand data to obtain a first data set;

[0021] Calculating the ratio of the photovoltaic power generation output power data to the computer CPU usage rate to obtain resource utilization, and generating a time feature according to the timestamp to obtain a second data set;

[0022] The second data set is subjected to time series processing by a lag term introduction method to obtain a processed data set.

[0023] In a possible implementation, the step of establishing a correlation model between photovoltaic power generation output power data and computer CPU computing power demand data based on the processed photovoltaic power generation output power data and the computer CPU computing power demand data includes:

[0024] Dividing the processed photovoltaic power generation output power data and the computer CPU computing power demand data into a training set and a test set;

[0025] Constructing a three-layer long short-term memory network LSTM time series model, wherein the three-layer long short-term memory network LSTM time series model includes: an input layer, an LSTM layer and an output layer;

[0026] The three-layer long short-term memory network LSTM time series model is trained by using the training set to obtain the trained three-layer long short-term memory network LSTM time series model;

[0027] The trained three-layer long short-term memory network LSTM timing model is tested and adjusted through the test set to obtain a model for associating the photovoltaic power generation output power data with the computer CPU computing power demand.

[0028] In a possible implementation, the step of testing and adjusting the trained three-layer long short-term memory network LSTM time series model through the test set to obtain the photovoltaic power generation output power data and the computer CPU computing power demand association model includes:

[0029] Input the test set into the trained three-layer long short-term memory network LSTM time series model to obtain the prediction result;

[0030] The model performance of the three-layer long short-term memory network LSTM time series model after training is evaluated by the indicator mean square error;

[0031] When the performance of the model is normal, a correlation model between photovoltaic power generation output power data and computer CPU computing power demand is obtained;

[0032] When the performance of the model is abnormal, the model input dimension is increased by increasing the number of LSTM layers and introducing more features, so as to obtain a model associating photovoltaic power generation output power data with computer CPU computing power requirements.

[0033] In a possible implementation manner, the function of the computer CPU usage and the computer CPU power consumption is constructed as a first formula; the first formula is:

[0034]

[0035] Among them, P is the computer CPU power consumption, f is the computer CPU utilization rate, and P0 is the initial value of CPU power consumption.

[0036] In a possible implementation, the step of regulating the computer CPU computing power according to the photovoltaic power generation output power data and the computer CPU computing power demand association model and the computer CPU usage rate and the computer CPU power consumption function includes:

[0037] Obtaining a predicted value of the computer CPU usage rate according to a correlation model between the photovoltaic power generation output power data and the computer CPU computing power demand;

[0038] Obtaining a CPU usage target value according to the computer CPU usage and the computer CPU power consumption function;

[0039] The CPU usage prediction value is smoothly adjusted so that the CPU usage prediction value gradually approaches the CPU usage target value.

[0040] In a possible implementation, the step of optimizing the photovoltaic power generation output power data and the computer CPU computing power demand association model, and the computer CPU usage rate and the computer CPU power consumption function according to the photovoltaic power generation output power data, the computer CPU computing power demand data and the computer CPU power consumption during operation includes:

[0041] Input the real-time computer CPU computing power demand data and the photovoltaic power generation output power data into the photovoltaic power generation output power data and computer CPU computing power demand association model, and continuously optimize the model;

[0042] The computer CPU usage rate and the computer CPU power consumption function are continuously optimized through the real-time computer CPU usage rate and the computer CPU power consumption.

[0043] In a possible implementation manner, the method for smoothly adjusting the predicted value of the computer CPU usage rate is to smoothly adjust the computer CPU usage rate through a PID controller, and the specific method is as follows:

[0044] PID input, calculate the deviation between the current computer CPU power consumption and the target computer CPU power consumption through the second formula; wherein the second formula is Δe=P 目标 -P, Δe is the deviation, P 目标 is the target computer CPU power consumption, P is the current computer CPU power consumption;

[0045] PID output, using the PID controller to output a computer CPU usage adjustment value to adjust the current computer CPU usage to approach the target computer CPU usage.

[0046] In a second aspect, the present application provides a distributed photovoltaic computer computing power control system, the system is used to execute the above-mentioned distributed photovoltaic computer computing power control method, the system comprising:

[0047] A data processing module is used to obtain photovoltaic power output data and computer CPU computing power demand data, and perform preprocessing; the computer CPU computing power demand data includes computer CPU usage rate, task queue length and memory usage;

[0048] A model building module, used to build a correlation model between photovoltaic power generation output power data and computer CPU computing power demand according to the processed photovoltaic power generation output power data and the computer CPU computing power demand data;

[0049] A function building module, used to build a function of the computer CPU usage rate and the computer CPU power consumption;

[0050] A computing power control module, which controls the computing power of the computer CPU according to the photovoltaic power generation output power data and the computer CPU computing power demand association model and the computer CPU usage rate and the computer CPU power consumption function;

[0051] The model optimization module optimizes the association model between the photovoltaic power generation output power data and the computer CPU computing power demand according to the photovoltaic power generation output power data, the computer CPU power consumption and the computer CPU resource usage during the operation.

[0052] The technical solution provided by this application may have the following beneficial effects:

[0053] Through the computer computing power control method and system of distributed photovoltaics in the present application, it is possible to establish a correlation model between photovoltaic power generation output power data and computer CPU computing power requirements, and construct a functional relationship between CPU usage and power consumption. Therefore, based on real-time monitoring of photovoltaic power generation and computer computing power requirements, and combined with a prediction algorithm, the computer's computing power allocation strategy can be adjusted, and the computer's CPU usage and computing task scheduling can be further dynamically adjusted, thereby maximizing the use of new energy and reducing the energy consumption of the computing system.

[0054] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0056] Figure 1 A flow chart showing a method for controlling computer computing power of distributed photovoltaics in an exemplary embodiment of the present disclosure;

[0057] Figure 2 A detailed flow chart showing step S100 of the computer computing power control method for distributed photovoltaics in an exemplary embodiment of the present disclosure is shown;

[0058] Figure 3 A detailed flow chart showing step S130 of the computer computing power control method for distributed photovoltaics in an exemplary embodiment of the present disclosure is shown;

[0059] Figure 4 A detailed flow chart showing step S200 of the computer computing power control method for distributed photovoltaics in an exemplary embodiment of the present disclosure is shown;

[0060] Figure 5 A detailed flow chart showing step S240 of the computer computing power control method for distributed photovoltaics in an exemplary embodiment of the present disclosure is shown;

[0061] Figure 6 A detailed flow chart showing step S400 of the computer computing power control method for distributed photovoltaics in an exemplary embodiment of the present disclosure is shown;

[0062] Figure 7 A detailed flow chart showing step S500 of the computer computing power control method for distributed photovoltaics in an exemplary embodiment of the present disclosure is shown;

[0063] Figure 8 An LSTM mapping relationship diagram showing the photovoltaic power output power data and the computer CPU computing power demand data of the distributed photovoltaic computer computing power control method in the exemplary embodiment of the present disclosure;

[0064] Fig. 9 A comparison diagram showing the computer computing power demand data of the prior art method for controlling the computer computing power of distributed photovoltaics in the exemplary embodiment of the present disclosure and the algorithm of the present application for controlling the computer computing power demand data;

[0065] Fig.10 A graph showing the relationship between the computer CPU usage rate and the computer CPU power consumption of the computer computing power control method of distributed photovoltaic in an exemplary embodiment of the present disclosure;

[0066] Fig.11A cluster analysis fitting diagram showing the computer CPU usage rate and computer CPU power consumption of the computer computing power control method of distributed photovoltaic in an exemplary embodiment of the present disclosure is shown;

[0067] Fig.12 A structural diagram of a computer computing power control system for distributed photovoltaics in an exemplary embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0068] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0069] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0070] In this exemplary embodiment, a distributed photovoltaic computer computing power control method is first provided, which can be applied to a terminal device, such as a mobile phone, a desktop computer, a personal digital assistant, a laptop computer, a tablet computer, a smart watch, or other mobile terminal. Figure 1 As shown in , the method may include the following steps:

[0071] Step S100: Acquire photovoltaic power output data and computer CPU computing power demand data, and perform preprocessing; the computer CPU computing power demand data includes computer CPU usage rate, task queue length and memory usage.

[0072] Step S200: establishing a photovoltaic power generation output power data and computer CPU computing power demand association model based on the processed photovoltaic power generation output power data and the computer CPU computing power demand data.

[0073] Step S300: constructing a function of the computer CPU usage and the computer CPU power consumption.

[0074] Step S400: regulating the computer CPU computing power according to the photovoltaic power generation output power data and the computer CPU computing power demand association model and the computer CPU usage rate and the computer CPU power consumption function.

[0075] Step S500: Optimizing the correlation model between the photovoltaic power generation output power data and the computer CPU computing power demand according to the photovoltaic power generation output power data, the computer CPU power consumption and the computer CPU resource usage during operation.

[0076] Through the above method, the photovoltaic power generation output power can be obtained through sensors and external interfaces, the real-time process of the computer computing power demand can be obtained by using the computer task manager, the photovoltaic power generation output power and the computer computing power demand information are handed over to the central processing unit for data preprocessing, and a correlation model between distributed photovoltaic output and computer CPU computing power demand is established. The functional relationship between CPU usage and power consumption is constructed, and the computer computing power is controlled according to the functional relationship and the correlation model. The CPU energy consumption parameters and the computer CPU resource usage are monitored, and the actual situation is fed back to the correlation model for optimization.

[0077] Next, we will refer to Figures 1 to 7 Each step of the above method in this example implementation is described in more detail.

[0078] In step S100, photovoltaic power generation output power data and computer CPU computing power demand data are obtained and preprocessed; the computer CPU computing power demand data includes computer CPU usage rate, task queue length and memory usage.

[0079] Optionally, the photovoltaic power generation output power is acquired through a sensor and a peripheral interface, and the sensor is connected to a peripheral interface of a computer for data transmission.

[0080] It should be noted that the measurement of photovoltaic output power is calculated by monitoring the output voltage and output current of the photovoltaic module. The output voltage is measured by the voltage sensor INA219 and outputs data through the I2C interface. The output current is measured by the Hall effect current sensor ACS712. The sensor converts the collected analog signal through the analog converter ADC to a digital signal for the processor to read. The combination of the two can obtain the instantaneous power of the photovoltaic output, which can be expressed as:

[0081] P(t)=V(t)×I(t)

[0082] Among them, is the output voltage, is the output current, and is the instantaneous output power.

[0083] In one embodiment, step S100 may include the following sub-steps:

[0084] In step S110, the output voltage and output current are collected by sensors arranged in the photovoltaic module to obtain the photovoltaic power generation output power data.

[0085] It should be noted that the collected power data is transmitted to the cloud server by wireless transmission and stored in the local database in real time for subsequent analysis and decision-making of the control algorithm.

[0086] Optionally, the sampling frequency of the collected data is set to collect data once every second. The data read by the processor after the sensor measurement may have errors. The voltage and current data are calibrated. The calibration formula is as follows:

[0087] V 校准 =V 测量 ×K voitaged

[0088] I 校准 =I 测量 ×K current

[0089] Among them, K voitaged is the voltage calibration coefficient, K current is the current calibration factor.

[0090] In step S120, the computer CPU computing power requirement data is obtained through the computer task manager.

[0091] It should be noted that in the Linux system, the psutil.cpu_percent() function is called to return the current CPU usage rate, and the interval parameter is set to automatically collect the CPU usage rate every second; psutil.virtual_memory() is called to obtain the system memory usage; psutil.process_iter() is called to traverse all processes in the current system and extract the PID, name, CPU usage rate and memory usage rate of each process. The collected data on the computer computing power requirements is stored in the SQLite database for subsequent analysis.

[0092] In step 130, the photovoltaic power generation output power data and the computer CPU computing power demand data are preprocessed to obtain a processed data set.

[0093] It should be noted that the distributed photovoltaic output power data preprocessing in the data preprocessing includes operations such as data cleaning and feature selection, and the preprocessing of computer CPU computing power demand data includes data standardization processing.

[0094] Furthermore, step S130 may include the following sub-steps:

[0095] In step S131, the missing values ​​and abnormal values ​​of the photovoltaic power generation output power data and the computer CPU computing power demand data are processed.

[0096] It should be noted that the data transmission is to transmit the photovoltaic output power stored on the remote server and the computer computing power demand data stored in the database to the central processing unit for processing using the Python requests library, and an interface is set in the central processing unit to obtain data from each data source. Step S131 is a data cleaning step, which uses a linear interpolation method to process missing data points during the collection process, and uses a box plot method to identify and process outliers within a set reasonable range.

[0097] In step S132, the photovoltaic power generation output power data is merged with the computer CPU computing power demand data to obtain a first data set.

[0098] It can be understood that, through the data fusion method, the photovoltaic output power and the CPU computing power data are fused to generate a unified data structure containing all relevant information.

[0099] In step S133, the ratio of the photovoltaic power generation output power data to the computer CPU usage rate is calculated to obtain resource utilization, and a time feature is generated according to the timestamp to obtain a second data set.

[0100] It can be understood that the above method extracts data feature values ​​and generates time features according to timestamps to perform periodic analysis.

[0101] In step S134, the second data set is subjected to time series processing by introducing a lag term to obtain a processed data set.

[0102] It should be noted that the preprocessed data will be stored in the central processing unit in a standardized format as a CSV file and transmitted to the upper-level algorithm for further processing and analysis; and the processed data will be monitored and visualized using Matplotlib and Seaborn libraries.

[0103] In step S200, a photovoltaic power generation output power data and a computer CPU computing power demand association model is established based on the processed photovoltaic power generation output power data and the computer CPU computing power demand data.

[0104] It should be noted that, the association model is constructed using a time series model, including data collection, data preprocessing, model selection and training, and model application. Data collection and data preprocessing are the method of step S100. After data preprocessing, data normalization is performed using Z-score standardization to scale the data to between [0,1] to eliminate the impact of different dimensions of photovoltaic power generation and computer computing power requirements.

[0105] In one embodiment, step S200 may include the following sub-steps:

[0106] In step S210, the processed photovoltaic power generation output power data and the computer CPU computing power requirement data are divided into a training set and a test set.

[0107] Exemplarily, the collected data is divided into a training set and a test set in a ratio of 7:3; the data is adjusted to a three-dimensional structure of LSTM input, namely [number of samples, time step, number of features].

[0108] In step S220, a three-layer long short-term memory network LSTM timing model is constructed, and the three-layer long short-term memory network LSTM timing model includes: an input layer, an LSTM layer and an output layer.

[0109] For example, the LSTM model is constructed using the Keras library of Python, including an input layer, an LSTM layer and an output layer. The first layer returns a sequence, the second layer does not return a sequence, and the output layer predicts the CPU computing power demand and the photovoltaic output power. The number of units in each LSTM layer is set to 50, the input dimension is determined to be (60, 2) according to the number of features, and the time step is set to 60. The mapping relationship diagram is as follows: Figure 8 shown.

[0110] In step S230, the three-layer long short-term memory network LSTM time series model is trained using the training set to obtain the trained three-layer long short-term memory network LSTM time series model.

[0111] For example, the LSTM model is trained using the training set, and the iteration rounds are set to 100 and the batch size is set to 64. At the same time, the training loss and validation loss are monitored during the training process to determine whether overfitting occurs.

[0112] In step S240, the trained three-layer long short-term memory network LSTM timing model is tested and adjusted through the test set to obtain a model for associating the photovoltaic power generation output power data with the computer CPU computing power demand.

[0113] It should be noted that, for example, predictions are made using the test set, and the prediction results are denormalized back to the original data range. The mean square error (MSE) indicator is used to evaluate the model performance, and finally a comparison chart of the predicted value and the true value is drawn to intuitively analyze the effect of the model. If the model performance is not ideal, the model input dimension can be increased by adjusting the parameters, increasing the number of LSTM layers, and introducing more features; after training, the model is saved and deployed to the production environment, and the real-time photovoltaic power output power data is used to input the LSTM model, and the CPU computing power demand and distributed photovoltaic output power are predicted in real time and the CPU usage rate is adjusted.

[0114] Further, step S240 may include the following sub-steps:

[0115] In step S241, the test set is input into the trained three-layer long short-term memory network LSTM time series model to obtain a prediction result.

[0116] In step S242, the model performance of the trained three-layer long short-term memory network LSTM time series model is evaluated by the indicator mean square error.

[0117] In step S243, when the model performance is normal, a model associating photovoltaic power generation output power data with computer CPU computing power requirements is obtained.

[0118] In step S244, when the model performance is abnormal, the model input dimension is increased by increasing the number of LSTM layers and introducing more features to obtain a correlation model between photovoltaic power generation output power data and computer CPU computing power demand.

[0119] It should be noted that if Fig. 9 As shown in the figure, the prediction of distributed photovoltaic output power through the LSTM model aims to better adjust the computer CPU computing power demand to achieve efficient use of energy, that is, when photovoltaic power generation is sufficient, increase the computing load and improve computing power; when photovoltaic power generation is insufficient, reduce computing power demand and reduce energy waste. At the same time, by real-time prediction of photovoltaic power generation and computer demand, dynamic scheduling and load management can be achieved, computer resources can be reasonably allocated, and the overall performance and response speed of the system can be improved.

[0120] In step S300, a function of the computer CPU usage and the computer CPU power consumption is constructed.

[0121] It is understandable that a functional relationship between CPU usage and computer CPU power consumption is constructed, and the computer computing power demand is regulated according to the LSTM model and the functional relationship.

[0122] In one embodiment, Figure 10-11As shown, the function of constructing the computer CPU usage and the computer CPU power consumption is a first formula; the first formula is:

[0123]

[0124] Among them, P is the computer CPU power consumption, f is the computer CPU utilization rate, and P0 is the initial value of CPU power consumption.

[0125] Furthermore, the power consumption control strategy is established by predicting the future photovoltaic output and computing power demand based on the LSTM model and combining the functional relationship between CPU frequency and power consumption:

[0126] Set a power consumption target value based on PV output to ensure that PV output matches power consumption requirements. In actual operation, monitor CPU power consumption in real time, compare actual power consumption with target power consumption value, and use the functional relationship between CPU frequency and power consumption to reversely solve the frequency f so that power consumption is close to target power consumption. The frequency formula can be expressed as:

[0127]

[0128] Among them, to ensure P target It can be solved only when it is greater than P0. The solved f target The CPU frequency that meets the target power consumption.

[0129] In step S400, the computer CPU computing power is regulated according to the photovoltaic power generation output power data and the computer CPU computing power demand association model and the computer CPU usage rate and the computer CPU power consumption function.

[0130] In one embodiment, step S400 may include the following sub-steps:

[0131] In step S410, the predicted value of the computer CPU usage rate is obtained according to the photovoltaic power generation output power data and the computer CPU computing power demand association model.

[0132] In step S420, a CPU usage target value is obtained according to the computer CPU usage and the computer CPU power consumption function.

[0133] In step S430, the CPU usage prediction value is smoothly adjusted so that the CPU usage prediction value gradually approaches the CPU usage target value.

[0134] For example, the CPU usage adjustment is to use a PID controller to smoothly adjust the CPU usage so that it gradually approaches f. target , the adjustment method is as follows:

[0135] PID input: Calculate the current computer CPU power consumption P and P target Deviation Δe=P target -P;

[0136] PID output: The PID controller is used to output a CPU usage adjustment value Δf to adjust the current CPU usage to approach the target CPU usage.

[0137] It should be noted that the power consumption control strategy uses the trained LSTM time series model, inputs the photovoltaic power output power and CPU computing power demand data at the current moment, and the model output predicts the future computing power demand. Based on the prediction results of the association model, combined with the relationship between CPU frequency and power consumption function, the system's CPU frequency and computing task priority are dynamically adjusted according to the photovoltaic output and the computer's computing power demand.

[0138] When the system runs multiple tasks, the computer tasks are prioritized. The computer task priority is to set the priorities of different computer tasks through the task manager, and divide the tasks into three categories: high priority, medium priority and low priority. Among them, high priority tasks represent critical tasks and must be run, even if the photovoltaic power generation is low; medium priority tasks represent normal tasks and run when the power generation is sufficient; low priority tasks represent non-critical tasks and are executed only when the power generation is sufficient; the present invention adopts a multi-level feedback queue scheduling algorithm combined with task priority scheduling;

[0139] High-priority task guarantee: When the photovoltaic output is insufficient, high-priority CPU usage is prioritized and low-priority CPU usage is reduced.

[0140] Low-priority tasks absorb photovoltaic power: When photovoltaic output is high, the low-priority CPU usage rate is increased to achieve higher power consumption, thereby better absorbing photovoltaic energy.

[0141] In step S500, according to the photovoltaic power generation output power data, computer CPU computing power demand data and computer CPU power consumption during operation, the photovoltaic power generation output power data and computer CPU computing power demand association model, and the computer CPU usage rate and computer CPU power consumption function are optimized.

[0142] In one embodiment, step S500 may include the following sub-steps:

[0143] In step S510, the real-time computer CPU computing power demand data and the photovoltaic power generation output power data are input into the photovoltaic power generation output power data and computer CPU computing power demand association model to continuously optimize the model.

[0144] It is understandable that during the regulation process, the operating status of the distributed photovoltaic output and the computer system is continuously monitored, and the actual operating results are fed back to the associated model; based on the feedback information, the associated model is optimized and adjusted to improve the accuracy and effectiveness of the regulation.

[0145] In step S520, the computer CPU usage rate and the computer CPU power consumption function are continuously optimized through the real-time computer CPU usage rate and the computer CPU power consumption.

[0146] It should be noted that the dynamic feedback control establishes a feedback closed loop, continuously monitors the actual power consumption and dynamically adjusts the frequency according to the deviation:

[0147] Power consumption feedback: Get the actual power consumption P of the system in real time and compare it with P 目标 Make comparisons;

[0148] Frequency fine-tuning: According to the power consumption deviation, the frequency 目标 Fine-tune the frequency nearby to ensure that the system always stays close to the target power consumption.

[0149] Furthermore, in this exemplary embodiment, a computer computing power control system for distributed photovoltaics is also provided. Fig.12 As shown in , the system may include:

[0150] A data processing module is used to obtain photovoltaic power output data and computer CPU computing power demand data, and perform preprocessing; the computer CPU computing power demand data includes computer CPU usage rate, task queue length and memory usage;

[0151] A model building module, used to build a correlation model between photovoltaic power generation output power data and computer CPU computing power demand according to the processed photovoltaic power generation output power data and the computer CPU computing power demand data;

[0152] A function building module, used to build a function of the computer CPU usage rate and the computer CPU power consumption;

[0153] A computing power control module, which controls the computing power of the computer CPU according to the photovoltaic power generation output power data and the computer CPU computing power demand association model and the computer CPU usage rate and the computer CPU power consumption function;

[0154] The model optimization module optimizes the association model between the photovoltaic power generation output power data and the computer CPU computing power demand according to the photovoltaic power generation output power data, the computer CPU power consumption and the computer CPU resource usage during the operation.

[0155] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0156] It should be noted that although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiment of the present disclosure, the features and functions of two or more modules or units described above can be concretized in one module or unit. Conversely, the features and functions of a module or unit described above can be further divided into multiple modules or units for concretization. The components displayed as modules or units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed scheme. Those of ordinary skill in the art can understand and implement it without paying creative work.

[0157] In an exemplary embodiment of the present disclosure, an electronic device is also provided, which may include a processor and a memory for storing executable instructions of the processor, wherein the processor is configured to execute the steps of the computer computing power control method for distributed photovoltaics described in any of the above embodiments by executing the executable instructions.

[0158] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as a system, method or program product. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as a "circuit", "module" or "system".

[0159] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the above-mentioned distributed photovoltaic computer computing power control method according to the implementation of the present disclosure.

[0160] In an exemplary embodiment of the present disclosure, a computer storage medium is also provided, on which a computer program is stored. When the program is executed by, for example, a processor, the steps of the computer computing power control method for distributed photovoltaics described in any of the above embodiments can be implemented.

[0161] In some possible embodiments, various aspects of the present invention may also be implemented in the form of a computer program product, which includes a computer program or instructions. When the computer program product is run on a terminal device, the computer program code or instructions are used to enable the terminal device to execute the steps of various exemplary embodiments of the present invention described in the above-mentioned distributed photovoltaic computer computing power control method section of this specification.

[0162] The program product described above may be written in any combination of one or more programming languages ​​to perform program code for the operation of the present invention, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0163] The computer software product may be stored in a computer storage medium, which may include a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable programmable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0164] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.

Claims

1. A computer computing power control method for distributed photovoltaics, characterized in that: include: Acquire photovoltaic power output data and computer CPU computing power demand data, and perform preprocessing; the computer CPU computing power demand data includes computer CPU usage rate, task queue length and memory usage; Establishing a correlation model between photovoltaic power generation output power data and computer CPU computing power demand data according to the processed photovoltaic power generation output power data and the computer CPU computing power demand data; Constructing a function of the computer CPU usage rate and the computer CPU power consumption; According to the photovoltaic power generation output power data and the computer CPU computing power demand association model and the computer CPU usage rate and the computer CPU power consumption function, the computer CPU computing power is regulated; According to the photovoltaic power generation output power data, computer CPU computing power demand data and computer CPU power consumption during operation, the photovoltaic power generation output power data and computer CPU computing power demand association model and the computer CPU utilization rate and computer CPU power consumption function are optimized.

2. The computer computing power control method for distributed photovoltaics according to claim 1 is characterized in that: The step of obtaining photovoltaic power generation output power data and computer CPU computing power demand data and performing preprocessing includes: The output voltage and output current are collected by sensors arranged in the photovoltaic modules to obtain the photovoltaic power generation output power data; Obtain computer CPU computing power demand data through the computer task manager; The photovoltaic power generation output power data and the computer CPU computing power demand data are preprocessed to obtain a processed data set.

3. The computer computing power control method for distributed photovoltaics according to claim 2 is characterized in that: The step of preprocessing the photovoltaic power generation output power data and the computer CPU computing power demand data to obtain a processed data set includes: Processing missing values ​​and abnormal values ​​of the photovoltaic power generation output power data and the computer CPU computing power demand data; The photovoltaic power generation output power data is merged with the computer CPU computing power demand data to obtain a first data set; Calculating the ratio of the photovoltaic power generation output power data to the computer CPU usage rate to obtain resource utilization, and generating a time feature according to the timestamp to obtain a second data set; The second data set is subjected to time series processing by a lag term introduction method to obtain a processed data set.

4. The computer computing power control method for distributed photovoltaics according to claim 1 is characterized in that: The step of establishing a correlation model between photovoltaic power generation output power data and computer CPU computing power demand data according to the processed photovoltaic power generation output power data and the computer CPU computing power demand data comprises: Dividing the processed photovoltaic power generation output power data and the computer CPU computing power demand data into a training set and a test set; Constructing a three-layer long short-term memory network LSTM time series model, wherein the three-layer long short-term memory network LSTM time series model includes: an input layer, an LSTM layer and an output layer; The three-layer long short-term memory network LSTM time series model is trained by using the training set to obtain the trained three-layer long short-term memory network LSTM time series model; The trained three-layer long short-term memory network LSTM timing model is tested and adjusted through the test set to obtain a model for associating the photovoltaic power generation output power data with the computer CPU computing power demand.

5. The computer computing power control method for distributed photovoltaics according to claim 4 is characterized in that: The step of testing and adjusting the trained three-layer long short-term memory network LSTM time series model through the test set to obtain the photovoltaic power generation output power data and the computer CPU computing power demand association model includes: Input the test set into the trained three-layer long short-term memory network LSTM time series model to obtain the prediction result; The model performance of the three-layer long short-term memory network LSTM time series model after training is evaluated by the indicator mean square error; When the performance of the model is normal, a correlation model between photovoltaic power generation output power data and computer CPU computing power demand is obtained; When the performance of the model is abnormal, the model input dimension is increased by increasing the number of LSTM layers and introducing more features, so as to obtain a model associating photovoltaic power generation output power data with computer CPU computing power requirements.

6. The computer computing power control method for distributed photovoltaics according to claim 1, characterized in that: The function of constructing the computer CPU usage rate and the computer CPU power consumption is a first formula; the first formula is: Among them, P is the computer CPU power consumption, f is the computer CPU utilization rate, and P0 is the initial value of CPU power consumption.

7. The computer computing power control method for distributed photovoltaics according to claim 6 is characterized in that: The step of regulating the computer CPU computing power according to the photovoltaic power generation output power data and the computer CPU computing power demand association model and the computer CPU usage rate and the computer CPU power consumption function includes: Obtaining a predicted value of the computer CPU usage rate according to a correlation model between the photovoltaic power generation output power data and the computer CPU computing power demand; Obtaining a CPU usage target value according to the computer CPU usage and the computer CPU power consumption function; The CPU usage prediction value is smoothly adjusted so that the CPU usage prediction value gradually approaches the CPU usage target value.

8. The computer computing power control method for distributed photovoltaics according to claim 4 is characterized in that: The step of optimizing the photovoltaic power generation output power data and the computer CPU computing power demand association model, the computer CPU usage rate and the computer CPU power consumption function according to the photovoltaic power generation output power data, the computer CPU computing power demand data and the computer CPU power consumption during operation includes: Input the real-time computer CPU computing power demand data and the photovoltaic power generation output power data into the photovoltaic power generation output power data and computer CPU computing power demand association model, and continuously optimize the model; The computer CPU usage rate and the computer CPU power consumption function are continuously optimized through the real-time computer CPU usage rate and the computer CPU power consumption.

9. The computer computing power control method for distributed photovoltaics according to claim 7, characterized in that: The method for smoothly adjusting the predicted value of the computer CPU usage rate is to smoothly adjust the computer CPU usage rate through a PID controller, and the specific method is as follows: PID input, calculate the deviation between the current computer CPU power consumption and the target computer CPU power consumption through the second formula; wherein the second formula is Δe=P 目标 -P, Δe is the deviation, P 目标 is the target computer CPU power consumption, P is the current computer CPU power consumption; PID output, using the PID controller to output a computer CPU usage adjustment value to adjust the current computer CPU usage to approach the target computer CPU usage.

10. A distributed photovoltaic computer computing power control system, characterized in that: The system is used to execute the computer computing power control method for distributed photovoltaics according to any one of claims 1 to 9, and the system comprises: A data processing module is used to obtain photovoltaic power output data and computer CPU computing power demand data, and perform preprocessing; the computer CPU computing power demand data includes computer CPU usage rate, task queue length and memory usage; A model building module, used to build a correlation model between photovoltaic power generation output power data and computer CPU computing power demand according to the processed photovoltaic power generation output power data and the computer CPU computing power demand data; A function building module, used to build a function of the computer CPU usage rate and the computer CPU power consumption; A computing power control module, which controls the computing power of the computer CPU according to the photovoltaic power generation output power data and the computer CPU computing power demand association model and the computer CPU usage rate and the computer CPU power consumption function; The model optimization module optimizes the association model between the photovoltaic power generation output power data and the computer CPU computing power demand according to the photovoltaic power generation output power data, the computer CPU power consumption and the computer CPU resource usage during the operation.

Images