Air conditioner energy efficiency optimization method based on big data

Abstract

The invention discloses an air conditioner energy efficiency optimization method based on big data, and belongs to the technical field of air conditioner energy efficiency optimization. The method solves the problems that in the prior art, the needed manufacturing cost is high, and the adaptability to the market is low, the threshold values of all the operation data of the air conditioner are obtained through the fuzzy reasoning technology and the neural network model, the energy consumption data comparison result is more accurate, a more accurate parameter basis is provided for air conditioner energy efficiency optimization, and the method is suitable for popularization and application. Therefore, the energy consumption parameters of the air conditioner are adjusted to be in a more appropriate state; through regular feedback of real-time operation data of the air conditioner and data detection, the operation pressure of an air conditioner optimization internal system is reduced on the premise that intelligent optimization of the energy consumption of the air conditioner is ensured; and after the parameters of the air conditioner are adjusted, the feedback and detection frequency is automatically modified, it is ensured that the operation state of the air conditioner is obtained at the first time, the situation that the energy consumption abnormity of the air conditioner reoccurs again is avoided to the maximum degree, and therefore the method has the practical value and the utilization prospect.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioning energy efficiency optimization, and in particular to an air conditioning energy efficiency optimization method based on big data. Background Art

[0002] At present, the development trend of the air-conditioning industry is generally characterized by market growth pressure, structural adjustment and technological innovation. In terms of structural adjustment, the air-conditioning market is transforming towards high-end and intelligentization. With consumers' pursuit of energy saving, health, comfort and smart life, air-conditioning products are also constantly upgrading; for example: high-efficiency and energy-saving products have increased their market share, and intelligent control functions have become standard. Technological innovation and brand competition are also important development trends in the air-conditioning industry; in the face of rising raw material costs, companies need to reduce costs and improve efficiency through technological innovation; in the future, as the global energy consumption trend gradually increases, the optimization of air-conditioning energy efficiency is imminent, aiming to improve the energy utilization efficiency of air-conditioning and reduce energy consumption, in order to cope with the "involution" phenomenon and achieve sustainable development.

[0003] In the existing technology, in order to achieve high energy efficiency, air-conditioning products usually adopt advanced technologies and materials, which leads to an increase in the manufacturing cost of the product, which is reflected in the market price, and then reduces the adaptability of the existing air-conditioning energy efficiency optimization method to the market.

[0004] Therefore, it does not meet the existing needs. We propose an air conditioning energy efficiency optimization method based on big data. Summary of the invention

[0005] The purpose of the present invention is to provide an air-conditioning energy efficiency optimization method based on big data, which obtains the threshold values ​​of various air-conditioning operation data through fuzzy reasoning technology and neural network model, makes the energy consumption data comparison result more accurate, and provides more accurate parameter basis for air-conditioning energy efficiency optimization, so as to adjust the air-conditioning energy consumption parameters to a more appropriate state; through the timed feedback of the real-time operation data and data detection of the air-conditioning, the operating pressure of the internal system of the air-conditioning optimization is reduced while ensuring the intelligent optimization of the air-conditioning energy consumption; and after adjusting the air-conditioning parameters, the frequency of feedback and detection is automatically modified to ensure that the operating status of the air-conditioning is obtained at the first time, and to avoid the recurrence of abnormal air-conditioning energy consumption to the greatest extent, so that the method has more practical value and utilization prospects, and solves the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for optimizing air conditioning energy efficiency based on big data, comprising the following steps:

[0008] Step 1: Collect the operating data of the air conditioner in real time through the sensor equipment. The operating data includes: historical and real-time operating data, specifically: temperature, humidity, wind speed, current, and voltage;

[0009] Step 2: Clean and pre-process the historical operation data to ensure its accuracy, and then use the historical operation data as a big data sample;

[0010] Step 3: Extract features from big data samples through fuzzy reasoning technology to identify key factors affecting air conditioning energy efficiency;

[0011] Step 4: Build a neural network model, use big data samples to train the neural network model, and obtain the thresholds of various operating data of the air conditioner under normal working conditions; where the air conditioner energy efficiency thresholds include: temperature threshold, humidity threshold, wind speed threshold, current threshold, and voltage threshold;

[0012] Step 5: The sensor device periodically feeds back the real-time collected air-conditioning operation data to the central processor, and compares the real-time air-conditioning operation data with the air-conditioning energy efficiency threshold to obtain the current energy efficiency state of the air-conditioning; formulates a corresponding optimization plan according to the current energy efficiency state of the air-conditioning, and the central processor adjusts the operation parameters of the air-conditioning in real time according to the corresponding optimization plan;

[0013] Step 6: After adjusting the operating parameters of the air conditioner, continue to collect the operating parameters of the air conditioner and feed them back to the central processing unit at regular intervals, and the central processing unit will perform energy efficiency detection again.

[0014] Furthermore, in step 1, the operation data, maintenance data and loss data of the air conditioner are collected in real time by the sensor device, which specifically includes the following steps:

[0015] Install the sensor equipment at key locations of the air conditioner, such as installing the wind speed sensor inside the air conditioner to ensure that it can accurately collect operating data, and use a screwdriver to fix the sensor to ensure that it is stable and reliable; then connect the sensor equipment to the air conditioner's control system to ensure that the data can be transmitted to the central processor in real time; set the sensor parameters and calibrate according to the specific model and requirements of the air conditioner to ensure the accuracy and reliability of the data; start the air conditioning system and collect the air conditioner's operating data in real time through the sensor equipment.

[0016] Furthermore, in step 2, the historical operation data is cleaned and preprocessed, which specifically includes the following steps:

[0017] Import data from different sources into a unified data environment, conduct preliminary exploratory analysis, and understand the structure, type, distribution, and potential problems of the data; detect missing values ​​in historical operation data, and select the mean processing method to fill them according to data characteristics and analysis, use statistical methods or visualization means to identify outliers in the data, and decide whether to remove, correct, or retain them; identify and delete duplicate data by comparing the similarity of records or unique identifiers to maintain the uniqueness of historical operation data; ensure consistency in data format, unit, naming, etc., and resolve conflicts and inconsistencies in historical operation data, such as: standardization of date formats and uniform case of strings; convert data into a form suitable for analysis, such as standardization, to ensure the uniformity of data formats.

[0018] Furthermore, in step three, feature extraction is performed on the big data sample by using fuzzy reasoning technology, which specifically includes the following steps:

[0019] A diagnostic matrix is ​​constructed based on experience and fuzzy statistics. The matrix includes: the degree of association between different symptoms and possible fault causes; the generalized hypothetical reasoning method is selected for fuzzy reasoning, and specific principles and methods of fuzzy reasoning are formulated to ensure the rationality and accuracy of the reasoning process; through fuzzy logic synthesis operations, the possible fault causes are comprehensively evaluated; based on the results of the synthesis operations, the possible fault causes are verified and the real fault causes are found.

[0020] Furthermore, before constructing the neural network model in step 4, the following steps are specifically included:

[0021] Obtain multiple groups of big data samples and organize them, and divide each group of big data samples into training set and test set in turn. The training set is used to train the neural network model and establish the feature layer of the neural network model; the test set is used to test the neural network model and verify the accuracy of the neural network model training results.

[0022] Furthermore, the neural network model is constructed in step 4, which specifically includes the following steps:

[0023] Construct a neural network model, select training sets and test sets of multiple groups of big data samples, import the training set into the neural network model for training, obtain the neural network model, and output the training result. The training result is: the preliminary threshold of air conditioning energy efficiency; then import the corresponding test set into the neural network model for testing, obtain the optimized neural network model, and output the test result. The test result is: the secondary threshold of air conditioning energy efficiency, compare the test result with the training result, and verify whether the preliminary threshold of air conditioning energy efficiency is consistent with the secondary threshold of air conditioning energy efficiency. If they are consistent, the air conditioning energy efficiency threshold is obtained; if not, retrain and test the neural network model until the two results are consistent.

[0024] Furthermore, the step 5 of periodically feeding back the air conditioning operation data collected in real time by the sensor device specifically includes the following steps:

[0025] The daily use time period of the air conditioner is collected, and the data is fed back at regular intervals according to the time period. The sensor device regularly feeds back data to the central processor according to the time node. The central processor regularly receives the real-time operation data of the air conditioner and stores it in chronological order to form an air conditioner database.

[0026] Furthermore, in step 5, the real-time operation data of the air conditioner is compared with the air conditioner energy efficiency threshold to obtain the current energy efficiency state of the air conditioner, which specifically includes the following steps:

[0027] After the central processing unit receives the real-time operation data of the air conditioner at regular intervals, it compares the real-time operation data of the air conditioner with the air conditioner energy efficiency threshold. If the real-time operation data of the air conditioner is within the air conditioner energy efficiency threshold range, it is determined that the current air conditioner energy efficiency is in a normal state; if the real-time operation data of the air conditioner is not within the air conditioner energy efficiency threshold range, it is determined that the current air conditioner energy efficiency is in an abnormal state, and the air conditioner operation parameters are adjusted in time according to the abnormal factors.

[0028] Furthermore, in step 6, after adjusting the operating parameters of the air conditioner, the operating parameters of the air conditioner are continuously collected and fed back to the central processor at regular intervals, and the central processor performs energy efficiency detection again, which specifically includes the following steps:

[0029] After adjusting the operating parameters of the air conditioner, the operating parameters of the air conditioner continue to be collected through the sensor equipment, and the data feedback time node is reset to increase the frequency of data feedback, shorten the time between each two data feedback time nodes, and adjust the number of energy efficiency detections according to the frequency of data feedback.

[0030] Furthermore, after continuing to collect the operating parameters of the air conditioner, the collected new data is cleaned and the data set is divided again, and the new data samples are imported into the neural network model for training and testing, which is used to optimize the characteristic parameters of the neural network model and timely adjust the air conditioner energy consumption threshold.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The present invention identifies key factors affecting the energy efficiency of the air conditioner through fuzzy reasoning technology, and uses a neural network model to obtain thresholds of various operating data of the air conditioner, so that the energy consumption data comparison result is more accurate, and a more accurate parameter basis is provided for the air conditioner energy efficiency optimization, thereby adjusting the air conditioner energy consumption parameters to a more appropriate state; through regular feedback of the real-time operating data and data detection of the air conditioner, while ensuring the intelligent optimization of the air conditioner energy consumption, the operating pressure of the air conditioner optimization internal system is reduced; and after the air conditioner parameters are adjusted, the feedback and detection frequencies are automatically modified to ensure that the operating status of the air conditioner is obtained at the first time, and to maximize the avoidance of the recurrence of abnormal air conditioner energy consumption, thereby making the method more practical and having more utilization prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flow chart of the air conditioning energy efficiency optimization method based on big data of the present invention. DETAILED DESCRIPTION

[0034] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] In order to solve the technical problem that in order to achieve high energy efficiency, air-conditioning products usually adopt advanced technology and materials, which leads to an increase in the manufacturing cost of the product, which is reflected in the market price and then reduces the applicability of the existing air-conditioning energy efficiency optimization method, please refer to Figure 1 , this embodiment provides the following technical solutions:

[0036] A method for optimizing air conditioning energy efficiency based on big data, comprising the following steps:

[0037] Step 1: Collect the operating data of the air conditioner in real time through sensor equipment. The operating data includes: historical and real-time operating data, specifically: temperature, humidity, wind speed, current, and voltage. The specific steps include: installing the sensor equipment at the key position of the air conditioner, such as: installing the wind speed sensor inside the air conditioner to ensure that it can accurately collect operating data, and use a screwdriver to fix the sensor to ensure that it is stable and reliable; then connect the sensor equipment to the control system of the air conditioner to ensure that the data can be transmitted to the central processing unit in real time; according to the specific model and requirements of the air conditioner, set the parameters of the sensor and calibrate it to ensure the accuracy and reliability of the data; start the air conditioning system, collect the operating data of the air conditioner in real time through the sensor equipment, and feed back the various operating data to the central processing unit.

[0038] In this embodiment, for example, prepare sensor devices, data cables and corresponding tools that are compatible with the air conditioner model, such as screwdrivers, wrenches, etc.; at the same time, it is necessary to understand the model and usage of the air conditioner in order to correctly install and configure the sensor; install the sensor device at the key position of the air conditioner, such as: the wind speed sensor is installed at the air inlet or outlet of the air conditioner to ensure that the wind speed data can be accurately collected; then use a screwdriver to fix the sensor, and connect the data cable to the control panel of the air conditioner or the external data acquisition device, that is, the central processing unit mentioned in this solution; according to the specific model and requirements of the air conditioner, configure the sampling frequency, data transmission method and other parameters of the sensor to ensure that the sensor can collect data in real time or at a fixed time, and transmit the data through the network or Bluetooth; after the sensor device starts running, it will automatically collect the operating data of the air conditioner, including but not limited to temperature, humidity, current, voltage, etc., so as to obtain the data parameters of the air conditioner energy efficiency.

[0039] Step 2: Clean and preprocess the historical operation data to ensure the accuracy of the historical operation data, and then use the historical operation data as a big data sample; specifically, the following steps are included:

[0040] Import data from different sources into a unified data environment, conduct preliminary exploratory analysis, and understand the structure, type, distribution, and potential problems of the data; detect missing values ​​in historical operation data, and select the mean processing method to fill them according to data characteristics and analysis, use statistical methods or visualization methods to identify outliers in the data, and decide whether to remove, correct, or retain them; identify and delete duplicate data by comparing the similarity of records or unique identifiers to maintain the uniqueness of historical operation data; ensure consistency in data format, unit, naming, etc., and resolve conflicts and inconsistencies in historical operation data, such as: standardization of date format, and uniform case of strings; convert data into a form suitable for analysis, such as: standardization to ensure the uniformity of data format; specifically, data cleaning can remove errors, missing values, duplicate values, and outliers in the data, thereby improving the quality and credibility of the data; by removing duplicate data and unnecessary information, the burden of data storage and processing can be reduced, the accuracy and reliability of analysis can be ensured, and misleading conclusions caused by dirty data can be reduced; in addition, cleaned data is easier to understand and process, which can speed up the data analysis and decision-making process.

[0041] Step 3: Extract features from big data samples through fuzzy reasoning technology to identify key factors affecting air conditioning energy efficiency; specifically, the following steps are included:

[0042] A diagnostic matrix is ​​constructed based on experience and fuzzy statistics. The matrix includes: the degree of association between different symptoms and possible fault causes; the generalized explanatory reasoning method is selected for fuzzy reasoning, and specific principles and methods of fuzzy reasoning are formulated to ensure the rationality and accuracy of the reasoning process; the possible fault causes are comprehensively evaluated through fuzzy logic synthesis operations; the possible fault causes are verified based on the results of the synthesis operations to find out the real fault causes; the determined fault causes are verified through actual inspection and testing to find out the real fault causes; it can help to effectively use fuzzy reasoning technology to extract feature data from air-conditioning operation data, thereby performing fault diagnosis and prediction.

[0043] Step 4: Construct a neural network model, use big data samples to train the neural network model, and obtain the thresholds of various operating data of the air conditioner under normal working conditions; wherein, the air conditioner energy efficiency threshold includes: temperature threshold, humidity threshold, wind speed threshold, current threshold, and voltage threshold; wherein, before constructing the neural network model, the following steps are specifically included: obtain multiple groups of big data samples and organize them, and divide each group of big data samples into training sets and test sets in turn, the training set is used to train the neural network model and establish the feature layer of the neural network model; the test set is used to test the neural network model and verify the accuracy of the training results of the neural network model; specifically, when dividing, the common division ratios are 70%-30%, 60%-40% or 8 0%-20%, and the consistency of data distribution should be maintained as much as possible during division, so the ratio of training set and test set can be set to 70%-30%; the bootstrap method is selected for division, and the training set and test set are generated by sampling with replacement. The specific steps are as follows: sampling with replacement is performed from the big data sample to generate a new data set as the training set, and the remaining unselected samples are used as the test set; the above process is repeated multiple times until a sufficient number of training sets and test sets are obtained; among them, since the air-conditioning operation data includes multiple types, when performing data division, it can be divided according to the data content, so that the training set and test set of each data can be exercised separately through the neural network model in the later stage, so as to obtain the threshold value of each operation parameter.

[0044] Building a neural network model includes the following steps:

[0045] Construct a neural network model, select training sets and test sets of multiple groups of big data samples, import the training set into the neural network model for training, obtain the neural network model, and output the training result, the training result is: the preliminary threshold of air conditioning energy efficiency; then import the corresponding test set into the neural network model for testing, obtain the optimized neural network model, and output the test result, the test result is: the secondary threshold of air conditioning energy efficiency, compare the test result with the training result, verify whether the preliminary threshold of air conditioning energy efficiency is consistent with the secondary threshold of air conditioning energy efficiency, if consistent, obtain the air conditioning energy efficiency threshold; if inconsistent, retrain and test the neural network model until the two results are consistent; in this embodiment, such as: select air conditioning The historical wind speed data samples are used for training to obtain the required air-conditioning operating temperature threshold; by obtaining the wind speed data set of the air-conditioning operation and determining the current season and temperature, such as winter and temperature between 0℃-5℃, these two factors are imported into the neural network model as characteristic factors affecting the air-conditioning wind speed, the wind speed training set is imported into the neural network model for training, and the training result is output as: ≤0.3m / s; the wind speed test set is then imported into the neural network model for testing, and the test result is output as: ≤0.3m / s; the two results are compared to conclude that the neural network model training process is more accurate, and ≤0.3m / s is regarded as the air-conditioning wind speed threshold, but this threshold is only for the air-conditioning wind speed under winter temperature.

[0046] Step 5: The sensor device periodically feeds back the real-time collected air-conditioning operation data to the central processor, and compares the real-time air-conditioning operation data with the air-conditioning energy efficiency threshold to obtain the current energy efficiency state of the air-conditioning; formulates a corresponding optimization plan according to the current energy efficiency state of the air-conditioning, and the central processor adjusts the operating parameters of the air-conditioning in real time according to the corresponding optimization plan; specifically, the following steps are included:

[0047] The daily use time period of the air conditioner is collected, and the data timing feedback time node is set according to the time period. The sensor device regularly feeds back data to the central processor according to the time node. The central processor regularly receives the real-time operation data of the air conditioner and stores it in time sequence to form an air conditioner database; in connection with the above embodiment, for example, by collecting the time period when users use the air conditioner daily in winter, which is mostly between 18:00pm and 8:00am, the data feedback time node can be set to feedback every 2 hours, such as 20:00pm, 22:00pm, 24:00pm, and so on; and then the transmission configuration of the sensor device is set, and its transmission time is set to the above-mentioned preset node, so that the sensor device feeds back data according to the specified time node.

[0048] After the central processing unit receives the real-time operation data of the air conditioner at a fixed time, the real-time operation data of the air conditioner is compared with the air conditioner energy efficiency threshold. If the real-time operation data of the air conditioner is within the air conditioner energy efficiency threshold, it is determined that the energy efficiency of the current air conditioner is in a normal state; if the real-time operation data of the air conditioner is not within the air conditioner energy efficiency threshold, it is determined that the energy efficiency of the current air conditioner is in an abnormal state, and the air conditioner operation parameters are adjusted in time according to the abnormal factors; Continuing with the above embodiment, for example: at 22:00 pm, the current air conditioner operation wind speed fed back by the sensor device to the central processing unit is 0 .8m / s, and compare it with the threshold value to get: 0.8m / s≥0.3m / s, from which it is known that the current air-conditioning wind speed is too high, the air-conditioning energy efficiency is in an abnormal state, and the air-conditioning wind speed parameters need to be adjusted; for example: at 24:00pm, the current air-conditioning operating wind speed fed back by the sensor device to the central processor is 0.25m / s, and compare it with the threshold value to get: 0.25m / s≤0.3m / s, from which it is known that the current air-conditioning wind speed is always, the air-conditioning energy efficiency is in a normal state, and there is no need to adjust the air-conditioning wind speed parameters.

[0049] Step 6: After adjusting the operating parameters of the air conditioner, continue to collect the operating parameters of the air conditioner and regularly feed them back to the central processor, which then performs energy efficiency testing again; specifically, the following steps are included:

[0050] After adjusting the operating parameters of the air conditioner, the operating parameters of the air conditioner are continuously collected through the sensor device, and the data feedback time node is reset to increase the frequency of data feedback, shorten the time between each two data feedback time nodes, and adjust the number of energy efficiency detections according to the frequency of data feedback; after continuing to collect the operating parameters of the air conditioner, the collected new data is re-cleaned and the data set is divided, and the new data samples are imported into the neural network model for training and testing, which is used to optimize the characteristic parameters of the neural network model and timely adjust the air conditioner energy consumption threshold; in accordance with the above embodiment, such as: feedback the current air conditioner at 22:00pm The operating wind speed is 0.8m / s. It is known that the current wind speed of the air conditioner is too high and the air conditioner wind speed parameters need to be adjusted. After reducing the wind speed, continue to collect wind speed parameters. At the same time, the original data feedback time interval is once every two hours, which can be adjusted to once every hour at this time. If the current feedback time node is 22:00pm, feedback can be made again at 23:00pm. If the feedback result shows normal, the original feedback time node is restored, that is, once every two hours. If the feedback result still shows abnormality, continue to feedback again in the next hour, that is, 24:00pm, until the feedback result shows normal.

[0051] The beneficial effects achieved by the above content are as follows: by constructing a neural network model, a more reasonable air-conditioning energy consumption threshold is obtained, which can achieve more accurate intelligent optimization; at the same time, by regularly feeding back real-time data, while ensuring the intelligent optimization of air-conditioning energy consumption, the operating pressure of the internal system of air-conditioning optimization is reduced, making this method more practical.

[0052] Working principle: The air conditioner's operating data is collected in real time through sensor equipment, and fuzzy reasoning technology is used to extract features from big data samples to identify key factors affecting the air conditioner's energy efficiency; the neural network model is trained using big data samples to obtain thresholds for various air conditioner operating data under normal working conditions; the sensor equipment regularly feeds back the real-time collected air conditioner operating data to the central processor, and compares the real-time air conditioner operating data with the air conditioner energy efficiency threshold to obtain the current air conditioner energy efficiency status; a corresponding optimization plan is formulated based on the current air conditioner's energy efficiency status, and the central processor adjusts the air conditioner's operating parameters in real time based on the corresponding optimization plan.

[0053] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0054] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and alterations may be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for optimizing air conditioning energy efficiency based on big data, characterized in that: The following steps are involved: Step 1: Collect the operating data of the air conditioner in real time through the sensor equipment. The operating data includes: historical and real-time operating data, specifically: temperature, humidity, wind speed, current, and voltage; Step 2: Clean and preprocess the historical operation data and use the historical operation data as a big data sample; Step 3: Extract features from big data samples through fuzzy reasoning technology to identify key factors affecting air conditioning energy efficiency; Step 4: Build a neural network model, use big data samples to train the neural network model, and obtain the thresholds of various operating data of the air conditioner under normal working conditions; where the air conditioner energy efficiency thresholds include: temperature threshold, humidity threshold, wind speed threshold, current threshold, and voltage threshold; Step 5: The sensor device periodically feeds back the real-time collected air-conditioning operation data to the central processor, and compares the real-time air-conditioning operation data with the air-conditioning energy efficiency threshold to obtain the current energy efficiency state of the air-conditioning; formulates a corresponding optimization plan according to the current energy efficiency state of the air-conditioning, and the central processor adjusts the operation parameters of the air-conditioning in real time according to the corresponding optimization plan; Step 6: After adjusting the operating parameters of the air conditioner, continue to collect the operating parameters of the air conditioner and feed them back to the central processing unit at regular intervals, and the central processing unit will perform energy efficiency detection again.

2. The air conditioning energy efficiency optimization method based on big data according to claim 1, characterized in that: In the step 1, the operation data, maintenance data and loss data of the air conditioner are collected in real time by the sensor device, which specifically includes the following steps: Install the sensor device at the key position of the air conditioner, and then connect the sensor device to the air conditioner control system; according to the specific model and requirements of the air conditioner, set the sensor parameters and calibrate it, start the air conditioning system, and collect the air conditioner operation data in real time through the sensor device.

3. The air conditioning energy efficiency optimization method based on big data according to claim 1, characterized in that: In step 2, the historical operation data is cleaned and preprocessed, which specifically includes the following steps: Detect missing values ​​in historical operation data, and select the mean processing method to fill them according to data characteristics and analysis. Use statistical methods or visualization means to identify outliers in the data and decide whether to remove, correct or retain them. Identify and delete duplicate data by comparing the similarity of records or unique identifiers, and then convert the data into a form suitable for analysis.

4. The air conditioning energy efficiency optimization method based on big data according to claim 1, characterized in that: In the step 3, feature extraction of the big data sample is performed by using fuzzy reasoning technology, which specifically includes the following steps: A diagnostic matrix is ​​constructed based on experience and fuzzy statistics. The matrix includes: the degree of association between different symptoms and possible fault causes; the generalized hypothetical reasoning method is selected for fuzzy reasoning, and specific principles and methods of fuzzy reasoning are formulated; through fuzzy logic synthesis operations, possible fault causes are comprehensively evaluated; based on the results of the synthesis operations, possible fault causes are verified and the real fault causes are found.

5. The air conditioning energy efficiency optimization method based on big data according to claim 1, characterized in that: Before constructing the neural network model in step 4, the following steps are specifically included: Obtain multiple groups of big data samples and organize them, and divide each group of big data samples into training set and test set in turn. The training set is used to train the neural network model and establish the feature layer of the neural network model; the test set is used to test the neural network model and verify the accuracy of the neural network model training results.

6. The air conditioning energy efficiency optimization method based on big data according to claim 5 is characterized in that: The neural network model is constructed in step 4, which specifically includes the following steps: Construct a neural network model, select training sets and test sets of multiple groups of big data samples, import the training set into the neural network model for training, obtain the neural network model, and output the training result. The training result is: the preliminary threshold of air conditioning energy efficiency; then import the corresponding test set into the neural network model for testing, obtain the optimized neural network model, and output the test result. The test result is: the secondary threshold of air conditioning energy efficiency, compare the test result with the training result, and verify whether the preliminary threshold of air conditioning energy efficiency is consistent with the secondary threshold of air conditioning energy efficiency. If they are consistent, the air conditioning energy efficiency threshold is obtained; if not, retrain and test the neural network model until the two results are consistent.

7. The air conditioning energy efficiency optimization method based on big data according to claim 1 is characterized in that: The step 5 of periodically feeding back the air conditioning operation data collected in real time by the sensor device specifically includes the following steps: The daily use time period of the air conditioner is collected, and the data is fed back at regular intervals according to the time period. The sensor device regularly feeds back data to the central processor according to the time node. The central processor regularly receives the real-time operation data of the air conditioner and stores it in chronological order to form an air conditioner database.

8. The air conditioning energy efficiency optimization method based on big data according to claim 7 is characterized in that: In step 5, the real-time operation data of the air conditioner is compared with the air conditioner energy efficiency threshold to obtain the current energy efficiency state of the air conditioner, which specifically includes the following steps: After the central processing unit receives the real-time operation data of the air conditioner at regular intervals, it compares the real-time operation data of the air conditioner with the air conditioner energy efficiency threshold. If the real-time operation data of the air conditioner is within the air conditioner energy efficiency threshold range, it is determined that the current air conditioner energy efficiency is in a normal state; if the real-time operation data of the air conditioner is not within the air conditioner energy efficiency threshold range, it is determined that the current air conditioner energy efficiency is in an abnormal state, and the air conditioner operation parameters are adjusted in time according to the abnormal factors.

9. The air conditioning energy efficiency optimization method based on big data according to claim 1, characterized in that: After adjusting the operating parameters of the air conditioner in step 6, the operating parameters of the air conditioner are continuously collected and fed back to the central processor at regular intervals, and the central processor performs energy efficiency detection again, which specifically includes the following steps: After adjusting the operating parameters of the air conditioner, the operating parameters of the air conditioner continue to be collected through the sensor equipment, and the data feedback time node is reset to increase the frequency of data feedback, shorten the time between each two data feedback time nodes, and adjust the number of energy efficiency detections according to the frequency of data feedback.

10. The air conditioning energy efficiency optimization method based on big data according to claim 9, characterized in that: After continuing to collect the operating parameters of the air conditioner, the collected new data is cleaned and the data set is divided again, and the new data samples are imported into the neural network model for training and testing to optimize the characteristic parameters of the neural network model and adjust the air conditioner energy consumption threshold in time.

Images