Energy-saving optimization control method and system, computer equipment and computer readable medium

Abstract

The invention relates to an energy-saving optimization control method and system, computer equipment and a computer readable medium. The method comprises the following steps: acquiring historical data of equipment; preprocessing the historical data; performing feature extraction on the preprocessed historical data; screening the extracted features to obtain target features representing the energy consumption of the equipment; constructing a neural network model based on the target features; performing energy consumption prediction on the equipment by using the neural network model; and regulating and controlling the equipment based on the energy consumption prediction result of the neural network model. According to the invention, the construction process can be simplified, and the prediction performance and calculation efficiency of the constructed neural network model can be improved; manual intervention can be reduced, the operation state of the equipment can be adjusted in real time according to environmental changes, energy waste can be avoided, the energy use efficiency and the regulation and control efficiency can be improved, and the optimal energy-saving effect can be achieved in time.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an energy-saving optimization control method, system, computer equipment and computer-readable medium. Background Art

[0002] Energy-saving control in parks and other places currently relies mainly on manual adjustment, and it is impossible to automatically adjust the operating status of the equipment. Manual adjustment cannot adjust the operating status of the equipment in real time according to environmental changes, and there is a delay in the adjustment. The equipment cannot achieve the optimal energy-saving effect in time, and there are problems such as energy waste and low control efficiency. Summary of the invention

[0003] Based on this, it is necessary to provide an energy-saving optimization control method, system, computer device and computer-readable medium to address the above technical problems.

[0004] In a first aspect, the present application provides an energy-saving optimization control method, the energy-saving optimization control method comprising:

[0005] Acquire historical data of a device, wherein the historical data includes historical environment data of an environment in which the device is located and historical working data of the device;

[0006] Preprocessing the historical data;

[0007] Extract features from the preprocessed historical data;

[0008] Screening the extracted features to obtain target features that characterize the energy consumption of the equipment;

[0009] Building a neural network model based on the target features;

[0010] Using the neural network model to predict energy consumption of the device;

[0011] Based on the energy consumption prediction result of the neural network model, the equipment is regulated.

[0012] Optionally, the preprocessing of the historical data includes:

[0013] Cleaning the historical data;

[0014] Normalize the cleaned historical data.

[0015] Optionally, the step of screening the extracted features to obtain target features characterizing the energy consumption of the device includes:

[0016] The correlation coefficient between each extracted feature and the energy consumption of the device is obtained based on the following formula:

[0017]

[0018] Among them, r xy represents the correlation coefficient between feature x and device energy consumption y, n is the total number of samples, x i is the characteristic value of the feature X of the i-th sample, is the mean of the feature X of n samples, y i is the equipment energy consumption of the i-th sample, is the mean of the equipment energy consumption of n samples;

[0019] Using a random forest model to evaluate the importance of the extracted features on the energy consumption of the device;

[0020] The extracted features are screened based on the correlation coefficient and the importance to obtain target features that characterize the energy consumption of the equipment.

[0021] Optionally, constructing a neural network model based on the target feature includes:

[0022] Select the initial neural network model;

[0023] Training the initial neural network model based on the target feature;

[0024] The learning rate of the trained initial neural network model is adjusted and regularized to obtain the neural network model.

[0025] Optionally, adjusting the learning rate of the trained initial neural network model includes: adjusting the learning rate of the trained initial neural network model using an adaptive learning rate method;

[0026] Regularizing the trained initial neural network model includes: introducing an L1 regularization term or an L2 regularization term into the trained initial neural network model.

[0027] Optionally, regulating the device based on the energy consumption prediction result of the neural network model includes:

[0028] Formulate a control strategy based on the energy consumption prediction results of the neural network model;

[0029] Generate a control instruction based on the control strategy;

[0030] The device is automatically regulated based on the regulation instruction.

[0031] Optionally, the control instruction includes an instruction for adjusting a working parameter of the device and a switch control instruction of the device; and automatically controlling the device based on the control instruction includes:

[0032] Dynamically adjust the operating parameters of the device based on the operating parameter adjustment instruction;

[0033] The device is controlled to be turned on or off based on the switch control instruction.

[0034] In a second aspect, the present application further provides an energy-saving optimization control system, the energy-saving optimization control system comprising:

[0035] A historical data acquisition module, used to acquire historical data of a device, wherein the historical data includes historical environment data of the environment in which the device is located and historical working data of the device;

[0036] A preprocessing module, used for preprocessing the historical data;

[0037] A feature extraction module is used to extract features from the preprocessed historical data;

[0038] A feature screening module is used to screen the extracted features to obtain target features that characterize the energy consumption of the equipment;

[0039] A neural network model building module, used to build a neural network model based on the target feature, and the neural network model is used to predict the energy consumption of the device;

[0040] A control module is used to control the device based on the energy consumption prediction result of the neural network model.

[0041] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any of the above energy-saving optimization control methods when executing the computer program.

[0042] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above energy-saving optimization control methods are implemented.

[0043] In the above-mentioned energy-saving optimization control method, system, computer device and computer-readable medium, the extracted features are screened after feature extraction to obtain target features that characterize the energy consumption of the equipment. When a neural network model is constructed based on the target features, since the target features are less than the extracted features, the construction process can be simplified and the prediction performance and computing efficiency of the constructed neural network model can be improved; by using the neural network model to predict the energy consumption of the equipment and regulating the equipment based on the energy consumption prediction results of the neural network model, human intervention can be reduced, the operating status of the equipment can be adjusted in real time according to environmental changes, energy waste can be avoided, energy utilization efficiency and regulation efficiency can be improved, and the optimal energy-saving effect can be achieved in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A flowchart of an energy-saving optimization control method provided in one embodiment of the present application;

[0045] Figure 2 A structural block diagram of an energy-saving optimization control system provided in another embodiment of the present application;

[0046] Figure 3 This is a diagram of the internal structure of a computer device in yet another embodiment of the present application. DETAILED DESCRIPTION

[0047] 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.

[0048] In one embodiment, see Figure 1 The present application provides an energy-saving optimization control method, which may include the following steps: S10 to S70.

[0049] S10: Acquire historical data of the device, where the historical data includes historical environment data of the environment in which the device is located and historical working data of the device.

[0050] S20: Preprocessing the historical data.

[0051] S30: Extract features from the preprocessed historical data.

[0052] S40: Screening the extracted features to obtain target features that characterize the energy consumption of the device.

[0053] S50: Constructing a neural network model based on the target features.

[0054] S60: Use the neural network model to predict energy consumption of the device.

[0055] S70: Regulating the device based on the energy consumption prediction result of the neural network model.

[0056] In the energy-saving optimization control method of the present application, the extracted features are screened after feature extraction to obtain target features that characterize the energy consumption of the equipment. When a neural network model is constructed based on the target features, since the target features are less than the extracted features, the construction process can be simplified and the prediction performance and computational efficiency of the constructed neural network model can be improved. By using the neural network model to predict the energy consumption of the equipment and regulating the equipment based on the energy consumption prediction results of the neural network model, human intervention can be reduced, the operating status of the equipment can be adjusted in real time according to environmental changes, energy waste can be avoided, energy utilization efficiency and regulation efficiency can be improved, and the optimal energy-saving effect can be achieved in a timely manner.

[0057] In step S10, refer to Figure 1 In step S10, historical data of the device is obtained, where the historical data includes historical environmental data of the environment in which the device is located and historical working data of the device.

[0058] As an example, a data acquisition module may be used to obtain historical data of the device; specifically, the data acquisition module may include a series of deployed high-precision sensors, and the data acquisition module may include a temperature sensor, a humidity sensor, a power monitoring sensor, and a device operation status monitoring sensor, etc. The data acquisition module maintains continuous and high-frequency data acquisition.

[0059] Specifically, a high-sensitivity temperature sensor can be used to capture the temperature change of the environment in which the device is located in real time, providing an important reference for subsequent energy consumption management.

[0060] Specifically, a humidity sensor may be used to monitor the fluctuation of relative humidity in the environment where the device is located, so as to ensure indoor comfort and prevent the device from being damaged by moisture.

[0061] Specifically, power monitoring sensors can be used to record the power usage of each device in the park in real time, providing data support for energy consumption analysis and energy saving of the equipment.

[0062] As an example, the data acquisition module may also include a meteorological data interface, which can obtain real-time meteorological data from a local weather station, such as wind speed, wind direction, and solar radiation information, to enhance the comprehensiveness and accuracy of environmental data.

[0063] In one embodiment, the above-mentioned various sensors can be installed at key locations in the park to ensure that all sensors can cover all monitoring areas. The sensor transmits the collected data to a storage device (such as a database) by wireless or wired means for storage as historical data for subsequent analysis and model training. It should be noted that before the data is transmitted to the storage device for storage, the data can also be preliminarily cleaned to remove obvious outliers.

[0064] In one implementation, taking a science and technology park with an area of ​​5,000 square meters as an example, 100 temperature sensors, 50 humidity sensors, 20 power monitoring sensors and 10 equipment operation status sensors can be deployed; these sensors can collect data every 5 minutes and transmit the data to the storage device in real time through the wireless network for storage.

[0065] As an example, the equipment may include all electrical equipment in the science and technology park, such as air conditioners and lighting equipment.

[0066] In step S20, refer to Figure 1 In step S20, the historical data is preprocessed.

[0067] As an example, step S20 may include the following steps: S201 - S202 .

[0068] S201: Cleaning the historical data.

[0069] S202: Normalize the cleaned historical data.

[0070] As an example, in step S201, the following steps may be included to process missing values, detect outliers, and process duplicate data on the historical data. In the process of missing value processing, the missing values ​​in the historical data may be identified, and the missing data may be filled with mean filling, median data filling, or a prediction model. In the process of outlier detection processing, outliers may be identified by methods such as box plots and Z-scores, and measures such as deletion, correction, or conversion may be taken to process the abnormal data. In the process of duplicate data processing, the uniqueness of each piece of data may be ensured by deleting or merging duplicate data records.

[0071] As an example, in step S202, a step of standardizing or normalizing the cleaned historical data may be included.

[0072] Specifically, in the process of data standardization, the cleaned historical data is scaled to a fixed range, such as between 0 and 1, which can be achieved by the following formula:

[0073]

[0074] Among them, X is the original historical data, X max is the maximum value in historical data, X min is the minimum value in the historical data.

[0075] Specifically, during the normalization process, the historical data can be normalized using methods such as logarithmic change or power transformation to reduce the skewness of data distribution.

[0076] In one implementation, taking the science and technology park as an example, we first received the original data set (i.e., historical data) from the data acquisition module. In the data cleaning stage, we identified and processed about 5% of missing values, deleted 2% of outliers, and merged 1% of duplicate data records. Subsequently, in the data normalization stage, we adopted a data standardization method to scale all historical environmental data and historical work data to the range of 0 to 1. After these processing, the data quality has been significantly improved, preparing for subsequent feature engineering and model training. In this way, we ensure that the data output by the data processing module is accurate, consistent, and useful, providing a reliable data source for the optimization of the energy-saving control system of the entire park.

[0077] In step S30, refer to Figure 1 In step S30, feature extraction is performed on the preprocessed historical data.

[0078] As an example, the temperature and humidity over a period of time can be extracted and analyzed to obtain the trend and periodicity of environmental changes. The time of the day can be divided into different time periods, such as morning, afternoon and evening, to consider the changing patterns of energy consumption in different time periods. The on / off status, operating hours and historical fault records of the equipment can be recorded, which is crucial for predicting energy consumption and equipment maintenance.

[0079] In step S40, refer to Figure 1 In step S40, the extracted features are screened to obtain target features that characterize the energy consumption of the device.

[0080] As an example, the extracted features can be filtered based on relevance and importance.

[0081] As an example, the correlation coefficient of each extracted feature for the energy consumption of the device may be obtained based on the following formula:

[0082]

[0083] Among them, r xy represents the correlation coefficient between feature x and device energy consumption y, n is the total number of samples, x i is the characteristic value of the feature X of the i-th sample, is the mean of the feature X of n samples, y i is the equipment energy consumption of the i-th sample, is the mean of the device energy consumption of n samples.

[0084] As an example, a random forest model may be used to evaluate the importance of the extracted features to the energy consumption of the device, thereby screening out features that contribute greatly to the subsequent prediction of the energy consumption of the device.

[0085] In one implementation, taking the science and technology park as an example, first, features such as temperature, humidity, time period, and equipment operating status are extracted from the acquired historical data; through correlation analysis, it can be found that temperature and humidity have a strong correlation with energy consumption. Next, the importance of the features is evaluated using a random forest model, and it can be found that temperature and equipment operating status contribute the most to the predicted equipment energy consumption. Ultimately, key features such as temperature, humidity, and equipment operating status can be selected as target features for subsequent neural network model training, thereby ensuring that the neural network model can accurately predict equipment energy consumption and provide data support for energy-saving control in the science and technology park. In this way, the feature dimension is reduced, and the prediction performance and computational efficiency of the subsequent neural network model are improved.

[0086] In step S50, refer to Figure 1 In step S50, a neural network model is constructed based on the target features.

[0087] As an example, step S50 may include the following steps: S501 to S503.

[0088] S501: Select an initial neural network model.

[0089] S502: Training the initial neural network model based on the target features.

[0090] S503: Adjusting the learning rate and regularizing the trained initial neural network model to obtain the neural network model.

[0091] Specifically, in step S501, in the selected initial neural network model, a multilayer perceptron (MLP) can be selected as the basic structure, which can capture the nonlinear complex relationship between features; ReLU (Rectified Linear Unit) is used as the activation function of the hidden layer to solve the gradient vanishing problem and improve the learning ability of the neural network model. The initial neural network model includes an input layer, two hidden layers and an output layer.

[0092] Specifically, in step S502, the target features can be divided into a training set, a validation set and a test set; wherein the training set is used to train the initial neural network model, the validation set is used to verify the trained initial neural network model, and the test set is used to test the verified initial neural network.

[0093] Specifically, in step S502, a back propagation (BP) algorithm may be used to update the weights. The BP algorithm adjusts the parameters of the initial neural network model by calculating the gradient of the loss function relative to each weight.

[0094] More specifically, the mean squared error (MSE) can be used as the loss function, and its formula can be as follows:

[0095]

[0096] Where n is the number of samples, Yi is the true value of the i-th sample, is the predicted value output by the initial neural network model.

[0097] Specifically, in step S503, an adaptive learning rate method (eg, Adam optimizer) may be used to adjust the learning rate of the trained initial neural network model to improve the stability and speed of the initial neural network model.

[0098] Specifically, in step S503, an L1 regularization term or an L2 regularization term may be introduced into the initial neural network model to regularize the initial neural network model. By introducing the L1 or L2 regularization term, overfitting can be prevented and the generalization ability of the obtained neural network model can be improved.

[0099] In step S60, refer to Figure 1 In step S60, the energy consumption of the device is predicted using the neural network model.

[0100] As an example, the method of obtaining the historical data of the device in step S10 can be continued to be used to obtain the real-time data of the device, and the real-time data can include the real-time environmental data of the environment in which the device is located and the real-time working data of the device. Specifically, a temperature sensor can be used to obtain the temperature of the environment in which the device is located in real time, and a humidity sensor can be used to obtain the humidity of the environment in which the device is located in real time. The obtained temperature and humidity can be used as the real-time environmental data of the device; a power detection sensor can be used to monitor the power usage of the device in real time, and a device operating status monitoring sensor can be used to monitor the device operating status of the device in real time. The obtained power usage and device operating status can be used as the real-time working data of the device.

[0101] As an example, the real-time data can be cleaned, normalized, and feature extracted, and the processed real-time data can be input into the neural network model to perform energy consumption prediction to obtain energy consumption prediction results. The energy consumption prediction results include predicted temperature values ​​(e.g., air conditioner temperature values) and predicted equipment operating states, etc.

[0102] In one embodiment, taking the science and technology park as an example, the target features can be first screened out, and an initial neural network model including an input layer, two hidden layers and an output layer can be constructed based on the target features. Through training, the obtained neural network model can accurately predict the temperature value that should be set for the air-conditioning system to maintain comfort and energy-saving efficiency in different time periods, as well as optimization suggestions for the equipment operating status. In the process of training the initial neural network model, the back propagation algorithm and the mean square error loss function can be used to optimize the parameters of the initial neural network model; through continuous iteration, the performance of the initial neural network model on the validation set is gradually improved, and finally a satisfactory prediction accuracy is achieved.

[0103] In step S70, refer to Figure 1 In step S70, the device is regulated based on the energy consumption prediction result of the neural network model.

[0104] As an example, step S70 may include the following steps: S701 to S703.

[0105] S701: Formulate a control strategy based on the energy consumption prediction results of the neural network model.

[0106] S702: Generate a control instruction based on the control strategy.

[0107] S703: Automatically control the device based on the control instruction.

[0108] Specifically, in step S702, the generated control instructions may include but are not limited to controlling the air-conditioning temperature, controlling the switch device, and the like.

[0109] Specifically, for the scheme of regulating the air conditioner temperature, in the regulation strategy in step S701, the operating temperature threshold of the air conditioner can be dynamically adjusted according to the energy consumption prediction result, so as to improve energy efficiency as much as possible while ensuring comfort. For example, the time of the day can be divided into multiple intervals, and different temperature thresholds are set for each interval to adapt to the changes in ambient temperature and the activity patterns of people in the science and technology park.

[0110] Specifically, for the switch device control solution, in the control strategy in step S701, the device can be turned on or off according to the operating status of the device in the energy consumption prediction result to avoid ineffective operation and reduce device energy consumption. At the same time, the operation of the device can be intelligently controlled in combination with factors such as the service life and maintenance status of the device to extend the life of the device and reduce maintenance costs.

[0111] As an example, the device may be automatically regulated based on the regulation instruction through a PLC system or a SCADA system.

[0112] As an example, the regulation instruction includes an operating parameter adjustment instruction of the device and a switch control instruction of the device; step S703 may include the following contents: dynamically adjusting the operating parameters of the device based on the operating parameter adjustment instruction; controlling the device to be turned on or off based on the switch control instruction.

[0113] In one embodiment, taking the science and technology park as an example, during the high temperature period in summer, based on the energy consumption prediction results of the neural network model, the air conditioning temperature threshold can be set at a level 1-2°C lower than normal based on the control instructions to reduce energy consumption. At night or on weekends when there are fewer people, the temperature threshold of the air conditioner is automatically adjusted to allow the indoor temperature to rise appropriately, or even turn off the air conditioner; at the same time, some lighting equipment can be turned off. Through the above-mentioned energy-saving optimization control method, it is possible to significantly reduce energy consumption without affecting the normal operation of the science and technology park. For example, within one month of implementing the energy-saving optimization control method, the overall operating efficiency of the science and technology park has been improved, which not only improves the energy utilization efficiency, but also provides intelligent and refined decision-making support for the management of the science and technology park.

[0114] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0115] In another embodiment, see Figure 2 The present application also provides an energy-saving optimization control system, which may include: a historical data acquisition module 10, a preprocessing module 20, a feature extraction module 30, a feature screening module 40, a neural network model construction module 50 and a control module 60; wherein the historical data acquisition module 10 is used to acquire the historical data of the device, and the historical data includes the historical environment data of the environment in which the device is located and the historical working data of the device; the preprocessing module 20 is used to preprocess the historical data; the feature extraction module 30 is used to extract features from the preprocessed historical data; the feature screening module 40 is used to screen the extracted features to obtain target features that characterize the energy consumption of the device; the neural network model construction module 50 is used to construct a neural network model based on the target features, and the neural network model is used to predict the energy consumption of the device; the control module 60 is used to control the device based on the energy consumption prediction result of the neural network model.

[0116] As an example, the energy-saving optimization control system can be used to perform the following steps: Figure 1 and the steps of the energy-saving optimization control method mentioned above; for the specific definition of the energy-saving optimization control system, please refer to the definition of the energy-saving optimization control method mentioned above, which will not be repeated here. Each module in the above energy-saving optimization control system can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0117] In another embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer 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 computer program in the non-volatile storage medium. The database of the computer device is used to store data such as deployment strategies. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it is implemented as follows Figure 1 And the energy-saving optimization control method described in the related embodiments.

[0118] In another embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it is implemented as follows Figure 1 The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0119] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0120] In another embodiment, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following is implemented: Figure 1 And the steps of the energy-saving optimization control method described in the relevant embodiments.

[0121] In another embodiment, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following Figure 1 And the steps of the energy-saving optimization control method described in the relevant embodiments.

[0122] 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 memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0123] 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.

[0124] The above-mentioned 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 invention patent. 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 patent of the present application shall be subject to the attached claims.

Claims

1. An energy-saving optimization control method, characterized in that: include: Acquire historical data of a device, wherein the historical data includes historical environment data of an environment in which the device is located and historical working data of the device; Preprocessing the historical data; Extract features from the preprocessed historical data; Screening the extracted features to obtain target features that characterize the energy consumption of the equipment; Building a neural network model based on the target features; Using the neural network model to predict energy consumption of the device; Based on the energy consumption prediction result of the neural network model, the equipment is regulated.

2. The energy-saving optimization control method according to claim 1, characterized in that: The preprocessing of the historical data includes: Cleaning the historical data; The cleaned historical data is normalized.

3. The energy-saving optimization control method according to claim 1, characterized in that: The extracted features are screened to obtain target features that characterize the energy consumption of the device, including: The correlation coefficient between each extracted feature and the energy consumption of the device is obtained based on the following formula: Among them, r xy represents the correlation coefficient between feature x and device energy consumption y, n is the total number of samples, x i is the characteristic value of the feature X of the i-th sample, is the mean of the feature X of n samples, y i is the equipment energy consumption of the i-th sample, is the mean value of the device energy consumption of n samples; Using a random forest model to evaluate the importance of the extracted features on the energy consumption of the device; The extracted features are screened based on the correlation coefficient and the importance to obtain target features that characterize the energy consumption of the equipment.

4. The energy-saving optimization control method according to claim 1, characterized in that: The step of constructing a neural network model based on the target feature comprises: Select the initial neural network model; Training the initial neural network model based on the target feature; The learning rate of the trained initial neural network model is adjusted and regularized to obtain the neural network model.

5. The energy-saving optimization control method according to claim 4, characterized in that: Adjusting the learning rate of the trained initial neural network model includes: adjusting the learning rate of the trained initial neural network model using an adaptive learning rate method; Regularizing the trained initial neural network model includes: introducing an L1 regularization term or an L2 regularization term into the trained initial neural network model.

6. The energy-saving optimization control method according to claim 1, characterized in that: The regulating the device based on the energy consumption prediction result of the neural network model includes: Formulate a control strategy based on the energy consumption prediction results of the neural network model; Generate a control instruction based on the control strategy; The device is automatically regulated based on the regulation instruction.

7. The energy-saving optimization control method according to claim 6, characterized in that: The control instructions include operating parameter adjustment instructions for the device and switch control instructions for the device; The automatically regulating the device based on the regulation instruction includes: Dynamically adjust the operating parameters of the device based on the operating parameter adjustment instruction; The device is controlled to be turned on or off based on the switch control instruction.

8. An energy-saving optimization control system, characterized in that: include: A historical data acquisition module, used to acquire historical data of a device, wherein the historical data includes historical environment data of the environment in which the device is located and historical working data of the device; A preprocessing module, used for preprocessing the historical data; A feature extraction module is used to extract features from the preprocessed historical data; A feature screening module is used to screen the extracted features to obtain target features that characterize the energy consumption of the equipment; A neural network model building module, used to build a neural network model based on the target feature, and the neural network model is used to predict the energy consumption of the device; A control module is used to control the device based on the energy consumption prediction result of the neural network model.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the energy-saving optimization control method according to any one of claims 1 to 7 are implemented.

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

Images