Air conditioner energy-saving data calculation and transmission method and system based on correlation verification

Abstract

The application discloses an air conditioner energy-saving data calculation and transmission method and system based on correlation verification, and relates to the field of air conditioner energy-saving data processing. The air conditioner energy-saving data calculation and transmission system based on correlation verification comprises a multi-source data acquisition module, a data verification module, an air conditioner energy-saving optimization module, a blockchain storage module and a storage verification module. The application collects multi-source air conditioner energy-saving data, verifies the data by using the correlation between different data and the time sequence change relationship of the data, ensures the accuracy of the collected data, avoids abnormal data caused by data collection errors, and calculates the energy-saving data of the air conditioner by using multi-dimensional data, thereby improving the credibility of the energy-saving data.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of air conditioning energy-saving data processing, and in particular to an air conditioning energy-saving data calculation and transmission method and system based on association verification. BACKGROUND

[0002] With the increasing global concern for energy efficiency and sustainable development, air conditioning systems, as one of the main energy-consuming devices, have gradually attracted attention for their energy-saving potential. With the popularization of smart home appliances and the development of Internet of Things technology, air conditioning systems have been able to achieve energy saving and emission reduction through intelligent control.

[0003] The intelligent control of air conditioners relies on data processing. The existing air conditioning energy-saving data processing method lacks data security and privacy. Traditional centralized data processing methods are vulnerable to external attacks, and the security of data and the privacy of users are difficult to guarantee. The data trustworthiness is insufficient, as the data may be tampered with during transmission and processing, resulting in doubts about the authenticity and accuracy of the final data. The efficiency is low, the traditional data processing process is complex and involves multiple intermediaries, resulting in low processing efficiency and high cost. There is a lack of energy-saving optimization mechanism, and data processing cannot effectively feedback control air conditioning equipment to optimize the energy-saving performance of air conditioners.

[0004] The introduction of blockchain technology in air conditioning energy-saving data processing solves the problems of data security and data authenticity in traditional air conditioning energy-saving data processing due to its decentralized, tamper-proof and transparent characteristics. However, the existing blockchain technology has the problem of low data processing capacity. SUMMARY

[0005] The present application proposes an air conditioning energy-saving data calculation and transmission method and system based on association verification to solve the problems of low blockchain data processing capacity, data security and energy-saving optimization in the prior art.

[0006] In order to achieve the above-mentioned purpose, the present application realizes the technical scheme as follows:

[0007] An air conditioning energy-saving data calculation and transmission method based on association verification, comprising:

[0008] S1. Collecting multi-source data of air conditioning equipment, including sensor data of air conditioner outdoor unit, sensor data of air conditioner indoor unit and multiple temperature sensor data of indoor space using air conditioner, the obtained data being a series of time series data;

[0009] S2. The multi-source data of the air conditioning equipment is sent to the data verification model for correlation verification and time sequence change relationship verification, the time sequence change relationship confidence and the correlation confidence of the to-be-processed energy-saving data are obtained, and the energy-saving confidence of the multi-source data of the air conditioning equipment is obtained after weighted calculation, the multi-source data of the air conditioning equipment is screened according to the energy-saving confidence, the credible energy-saving data is obtained, the energy-saving value of the air conditioning equipment is calculated using the credible energy-saving data, the data eliminated in the screening process is taken as the energy-saving abnormal data, and the energy-saving abnormal data is marked and recorded;

[0010] The calculation formula of the energy-saving value is:

[0011] ;

[0012] In the formula, D j represents the jth energy-saving data, C j represents the energy-saving confidence of the jth energy-saving data, M j represents the maximum credible value of the jth energy-saving data, M li represents the ith eliminated energy-saving data, and γ represents the weight of the eliminated data.

[0013] S3. The credible energy-saving data is sent to the energy-saving optimization model for optimization identification, and the optimization identification result is obtained, and the air conditioning equipment is subjected to energy-saving suggestion and energy-saving optimization strategy adjustment through the optimization identification result.

[0014] S3. The credible energy-saving data is sent to the energy-saving optimization model for optimization identification, and the optimization identification result is obtained, and the air conditioning equipment is subjected to energy-saving suggestion and energy-saving optimization strategy adjustment through the optimization identification result.

[0015] S4. The multi-source data of the air conditioning equipment is split using a random splitting method, the split data is encrypted using a multi-layer encryption algorithm to obtain energy-saving data ciphertext, different energy-saving data ciphertexts are stored in different addresses, the multi-source data of the air conditioning equipment is subjected to simplified identification processing to obtain simplified identifications of different data of the air conditioning equipment, all simplified identifications of the air conditioning equipment are associated with the energy-saving value to obtain air conditioning energy-saving monitoring data, the air conditioning energy-saving monitoring data is uploaded to the blockchain, the air conditioning energy-saving monitoring data is verified through the energy-saving smart contract of the blockchain, and the verified air conditioning energy-saving monitoring data is stored in the blockchain.

[0016] S5. Two energy-saving data ciphertexts of the energy-saving value of one air conditioning equipment are randomly obtained, the plaintext is obtained after decryption, the plaintext is compared with the simplified identification in the air conditioning energy-saving monitoring data, the data is verified by comparing the plaintext with the plaintext.

[0017] As a preferred technical solution of the present invention, the data verification model includes a data processing layer, a time series change relationship verification layer, an association relationship verification layer, and a confidence acquisition layer;

[0018] The data processing layer is used to preprocess the acquired multi-source data to obtain the energy-saving data to be processed;

[0019] The temporal relationship verification layer is used to verify the temporal change relationship of the energy-saving data to be processed. By the pattern differences in the temporal change behavior of the data, the confidence of the temporal change relationship of the energy-saving data to be processed is obtained, and the temporal feature data of the energy-saving data to be processed is generated.

[0020] The correlation verification layer is used to verify the correlation between time-series feature data from different sources. By comparing the correlation between data with the standard correlation, the confidence level of the correlation of the energy-saving data to be processed is obtained.

[0021] The confidence level acquisition layer is used to perform weighted calculations on the confidence levels of time-series change relationships and association relationships to obtain the energy-saving confidence levels of each data point in the multi-source data. The formula for calculating the energy-saving confidence level is as follows:

[0022] ;

[0023] in the formula These represent the confidence scores for time-series changes and the association relationships, respectively. These are the validation weights for the confidence scores of time-series change relationships and association relationships, respectively, α. a +α b =1, 0.6≥α a ≥0.2, 0.8≥α b ≥0.2.

[0024] As a preferred embodiment of the present invention, it further includes constructing a data verification model, the specific steps of which are as follows:

[0025] A1. Obtain multi-source data from several air conditioning devices, perform preprocessing operations on the multi-source data and label the confidence level of time-series change relationships and the confidence level of association relationships, obtain the time-series feature data of the preprocessed multi-source data, combine the preprocessed and labeled multi-source data and the time-series feature data to form a data validation dataset, and divide the data validation dataset into a data validation training set and a data validation test set;

[0026] A2. The data validation training set is fed into the initial time series validation model and the initial relationship validation model built based on deep neural networks for training. The optimized time series validation model and the optimized relationship validation model are obtained by taking the time series change relationship confidence label and the association relationship confidence as the objectives, respectively.

[0027] A3. The output verification test set is sent into the optimized timing verification model and the optimized relationship verification model for accuracy evaluation, respectively taking the timing change relationship confidence label and the correlation relationship confidence as the target, adjusting the hyperparameters of the optimized timing verification model and the optimized relationship verification model until the accuracy is within the error range, and finally obtaining the optimized timing verification model and the optimized relationship verification model as the timing change relationship verification layer and the correlation relationship verification layer of the data verification model;

[0028] A4. The verification weights of the timing change relationship confidence label and the correlation relationship confidence label are obtained through the timing characteristics and the data relationship quantity of the multi-source data, and the energy saving confidence of each data in the multi-source data is obtained by weighted calculation of the timing change relationship confidence and the correlation relationship confidence according to the verification weights.

[0029] As a preferred technical scheme of the present application, the energy saving optimization model comprises a data processing layer, an optimization identification layer and an optimization suggestion layer.

[0030] The data processing layer is used for pre-processing the trusted energy saving data to obtain the to-be-identified energy saving data.

[0031] The optimization identification layer is used for identifying the to-be-identified energy saving data to obtain the optimization identification result.

[0032] The optimization suggestion layer is used for matching the optimization adjustment strategy in the optimization strategy library according to the optimization identification result, and generating the corresponding energy saving suggestion.

[0033] As a preferred technical scheme of the present application, it further comprises constructing an energy saving optimization model, and the specific steps are as follows:

[0034] B1. Obtain the multi-source data of a plurality of air conditioning equipment, pre-process the multi-source data and mark the optimization identification result label, group the multi-source data after pre-processing and marking the label into an energy saving optimization data set, and divide the energy saving optimization data set into an energy saving optimization training set and an energy saving optimization test set.

[0035] B2. The energy saving optimization test set is sent into the initial energy saving optimization model constructed based on the SVM model for training, taking the optimization identification result label as the target, to obtain a strengthened energy saving optimization model.

[0036] B3. The energy saving optimization test set is sent into the strengthened energy saving optimization model for accuracy evaluation, taking the optimization identification result label as the target, adjusting the hyperparameters of the strengthened energy saving optimization model until the accuracy is within the error range, and finally obtaining the strengthened energy saving optimization model as the optimization identification layer of the energy saving optimization model.

[0037] B4. The standard optimization adjustment strategy is matched in the optimization strategy library through the optimization identification result label, the matched standard optimization adjustment strategy is taken as the output, and the corresponding energy saving suggestion is output.

[0038] As a preferred technical solution of the present application, the multi-source data of the air conditioner device is split, and part of the multi-source data is randomly selected and removed, the amount of the part of data being obtained by a random function, and the upper limit of the part of data being less than half of the amount of data.

[0039] As a preferred technical solution of the present application, the split data is encrypted using a multi-layer encryption algorithm to obtain energy-saving data ciphertext, and the specific steps of encryption and decryption are as follows:

[0040] Data encryption:

[0041] C1. The original data is encrypted using an AES key A to obtain ciphertext C.

[0042] C2. The AES key A is encrypted using an RSA public key to obtain an AES key hash value h of a fixed length.

[0043] C3. The ciphertext C and the AES key hash value h are combined to form a ciphertext file T.

[0044] Data decryption:

[0045] C4. The ciphertext C and the AES key hash value h are extracted from the ciphertext file T.

[0046] C5. The AES key A is obtained by decrypting the AES key hash value h using an RSA private key.

[0047] C6. The original data is obtained by decrypting the ciphertext C using the AES key A.

[0048] As a preferred technical solution of the present application, the energy-saving smart contract includes energy-saving data owner verification, data submission time verification, and energy-saving location verification.

[0049] An air conditioner energy-saving data calculation and transmission system based on association verification includes:

[0050] A multi-source data acquisition module is configured to acquire multi-source data of an air conditioner device, including sensor data of an air conditioner outdoor unit, sensor data of an air conditioner indoor unit, and data of multiple temperature sensors in a room using the air conditioner.

[0051] A data verification module is configured to input the multi-source data of the air conditioner device into a data verification model for association relationship verification and time sequence change relationship verification, obtain an energy-saving confidence of the multi-source data of the air conditioner device, and filter the multi-source data of the air conditioner device according to the energy-saving confidence to obtain trusted energy-saving data, calculate an energy-saving value of the air conditioner device using the trusted energy-saving data, and remove data as energy-saving abnormal data during the filtering process and mark and record the energy-saving abnormal data.

[0052] The air conditioner energy-saving optimization module is used for sending trusted energy-saving data into an energy-saving optimization model for optimization identification, obtaining an optimization identification result, and making energy-saving suggestions and adjusting energy-saving optimization strategies for the air conditioner equipment through the optimization identification result.

[0053] The blockchain storage module is used for splitting the multi-source data of the air conditioner equipment using a random splitting method, encrypting the split data using a multi-layer encryption algorithm to obtain energy-saving data ciphertext, storing different energy-saving data ciphertext in different addresses, simplifying the identification of the multi-source data of the air conditioner equipment to obtain the simplified identification of different data of the air conditioner equipment, associating all the simplified identification of the air conditioner equipment with energy-saving values to obtain air conditioner energy-saving monitoring data, uploading the air conditioner energy-saving monitoring data to the blockchain, verifying the air conditioner energy-saving monitoring data through the energy-saving smart contract of the blockchain, and storing the verified air conditioner energy-saving monitoring data in the blockchain.

[0054] The storage verification module is used for randomly obtaining two energy-saving data ciphertext of the energy-saving value of one air conditioner equipment, decrypting the ciphertext to obtain plaintext, comparing the plaintext with the simplified identification in the air conditioner energy-saving monitoring data, comparing the plaintext with the plaintext, and verifying the data.

[0055] The present application has the following advantages:

[0056] 1. The present application collects multi-source air conditioner energy-saving data, obtains the time sequence change relationship confidence of the to-be-processed energy-saving data according to the time sequence change relationship of the data itself, obtains the association relationship confidence of the data according to the association relationship between different data, and obtains the energy-saving confidence of the multi-source data of the air conditioner equipment after weighted calculation, so as to ensure the accuracy of the collected data, avoid abnormal data caused by error in data collection, improve the credibility of the energy-saving data, and calculate the energy-saving value of the air conditioner through multi-source data and corresponding energy-saving confidence. The original data is weighted and calculated through the energy-saving confidence. In the calculation process, the removed data is calculated by the energy-saving confidence. The obtained energy-saving value fully uses all the data on the basis of credibility, and realizes the credible calculation of the energy-saving value in the case of data loss.

[0057] 2. The present application analyzes multi-source data, makes energy-saving suggestions and energy-saving optimization strategy adjustments for the air conditioner according to the analysis result, and improves the dynamic adjustment ability of air conditioner energy-saving optimization.

[0058] 3. The present application stores the energy-saving data of the air conditioner by using the blockchain combined with the distributed storage mode, prevents the security of the stored energy-saving data through the consensus mechanism of the blockchain, and improves the standardization of the energy-saving data through the energy-saving smart contract of the blockchain.

[0059] 4、The air conditioner energy-saving monitoring data is obtained by compressing and encrypting the energy-saving data, so that the small amount of air conditioner energy-saving monitoring data can be efficiently stored in the block chain, and the energy-saving value of the air conditioner equipment and the simplified identification are associated to verify the air conditioner energy-saving monitoring data, so that the credibility of the data is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 A structure diagram of an air conditioner energy-saving data calculation and transmission system based on association verification used by an embodiment of the present application;

[0061] Figure 2 A flowchart of an air conditioner energy-saving data calculation and transmission method based on association verification used by an embodiment of the present application. DETAILED DESCRIPTION

[0062] In order to enable personnel in the technical field to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application.

[0063] Embodiment 1, an air conditioner energy-saving data calculation and transmission method based on association verification, as shown in Figure 2 , includes:

[0064] S1. Collecting multi-source data of air conditioner equipment, including sensor data of air conditioner outdoor unit, sensor data of air conditioner indoor unit and multiple temperature sensor data of indoor space using air conditioner, and obtaining a series of time sequence data;

[0065] S2. Sending the multi-source data of air conditioner equipment into a data verification model for association relationship verification and time sequence change relationship verification, obtaining energy-saving confidence of the multi-source data of air conditioner equipment, and filtering the multi-source data of air conditioner equipment according to the energy-saving confidence, obtaining credible energy-saving data, calculating the energy-saving value of air conditioner equipment using the credible energy-saving data, and discarding the data in the filtering process as energy-saving abnormal data, and marking and recording the energy-saving abnormal data;

[0066] The purpose of marking and recording the energy-saving abnormal data is to use these data for abnormal feedback;

[0067] Discarding the data with energy-saving confidence less than a preset threshold in the multi-source data of air conditioner equipment, the preset threshold is configured by historical credit evaluation degree of air conditioner owner, and the remaining data is used as credible energy-saving data, and the calculation formula of energy-saving value is:

[0068] ;

[0069] In the formula, D j represents the jth energy-saving data, C jM represents the energy-saving confidence of the jth energy-saving data j M represents the maximum confidence value of the jth energy-saving data li γ represents the weight of the eliminated energy-saving data, and i represents the ith eliminated energy-saving data The value of the formula decreases as the historical credit evaluation degree of the air conditioner owner decreases; the formula can fully combine the obtained full data to calculate the energy-saving value, so that the energy-saving value obtained in the case of data loss is still reliable;

[0070] The data verification model includes a data processing layer, a time sequence change relationship verification layer, an association relationship verification layer, and a confidence acquisition layer.

[0071] The data processing layer is used for pre-processing the obtained multi-source data to obtain the to-be-processed energy-saving data.

[0072] The time sequence relationship verification layer is used for verifying the time sequence change relationship of the to-be-processed energy-saving data, obtaining the time sequence change relationship confidence of the to-be-processed energy-saving data through the mode difference of the data in the time sequence change behavior, and generating the time sequence feature data of the to-be-processed energy-saving data.

[0073] The association relationship verification layer is used for verifying the association relationship between the time sequence feature data of different source data, and obtaining the association relationship confidence of the to-be-processed energy-saving data through the difference between the association relationship between the data and the standard association relationship.

[0074] The confidence acquisition layer is used for weighted calculation of the time sequence change relationship confidence and the association relationship confidence, to obtain the energy-saving confidence of each data in the multi-source data, and the calculation formula of the energy-saving confidence is:

[0075] ;

[0076] In the formula, M represents the energy-saving confidence of the jth energy-saving data respectively represent the time sequence change relationship confidence and the association relationship confidence, respectively represent the verification weights of the time sequence change relationship confidence and the association relationship confidence, and α a + α b = 1, 0.6≥ α a ≥ 0.2, 0.8≥ α b ≥ 0.2; the values of the two verification weights are respectively determined by the time sequence data standard degree and the data standard degree of the multi-source data, the time sequence data standard degree depends on the proportion of the frequency of collecting time sequence data to the standard frequency, and the data standard degree depends on the proportion of the number of multi-source data to the number of standard multi-source data, and the closer the value is to 100%, the larger it is. Specifically, when the proportion is 100%, the maximum value is taken. Through this dynamic weight setting, the energy-saving confidence calculation can be more flexible in different data situations in the specific implementation process, and the robustness of the energy-saving confidence is improved.

[0077] Further comprising constructing a data verification model, the specific steps are:

[0078] A1. Obtain multi-source data of a plurality of air conditioning equipment, pre-process the multi-source data and label the time sequence change relationship confidence label and the correlation relationship confidence, obtain the time sequence feature data of the pre-processed multi-source data, group the pre-processed multi-source data and the time sequence feature data into a data verification dataset, and divide the data verification dataset into a data verification training set and a data verification test set;

[0079] A2. The data verification training set is sent into the initial time sequence verification model and the initial relationship verification model based on the deep neural network for training, and the time sequence change relationship confidence label and the correlation relationship confidence are taken as the target respectively, and the optimized time sequence verification model and the optimized relationship verification model are obtained;

[0080] A3. The output verification test set is sent into the optimized time sequence verification model and the optimized relationship verification model for accuracy evaluation, and the time sequence change relationship confidence label and the correlation relationship confidence are taken as the target respectively, and the hyperparameters of the optimized time sequence verification model and the optimized relationship verification model are adjusted until the accuracy is within the error range, and the finally obtained optimized time sequence verification model and optimized relationship verification model are taken as the time sequence change relationship verification layer and the correlation relationship verification layer of the data verification model;

[0081] A4. The verification weights of the time sequence change relationship confidence label and the correlation relationship confidence label are obtained through the time sequence features and the data relationship quantities of the multi-source data, and the energy saving confidence of each data in the multi-source data is obtained by weighted calculation of the time sequence change relationship confidence and the correlation relationship confidence according to the verification weights.

[0082] S3. The trusted energy saving data is sent into the energy saving optimization model for optimization identification, and the optimization identification result is obtained, and the energy saving suggestion and the energy saving optimization strategy adjustment are made to the air conditioning equipment through the optimization identification result;

[0083] The energy saving optimization model comprises a data processing layer, an optimization identification layer and an optimization suggestion layer;

[0084] The data processing layer is used for pre-processing the trusted energy saving data to obtain the to-be-identified energy saving data;

[0085] The optimization identification layer is used for identifying the to-be-identified energy saving data to obtain the optimization identification result;

[0086] The optimization suggestion layer is used for matching the optimization adjustment strategy in the optimization strategy library according to the optimization identification result, and generating the corresponding energy saving suggestion.

[0087] Further comprising constructing an energy saving optimization model, the specific steps are:

[0088] B1. Obtain multi-source data of a plurality of air conditioning devices, pre-process the multi-source data, and label optimization identification result tags, group the pre-processed and labeled multi-source data into an energy-saving optimization dataset, and divide the energy-saving optimization dataset into an energy-saving optimization training set and an energy-saving optimization test set;

[0089] B2. Train the energy-saving optimization test set in an initial energy-saving optimization model based on an SVM model to optimize the identification result tags, and obtain a reinforced energy-saving optimization model;

[0090] B3. Evaluate the accuracy of the energy-saving optimization test set in the reinforced energy-saving optimization model, adjust the hyperparameters of the reinforced energy-saving optimization model until the accuracy is within the error range, and finally obtain the reinforced energy-saving optimization model as the optimization identification layer of the energy-saving optimization model;

[0091] B4. Match the standard optimization adjustment strategy in the optimization strategy library through the optimization identification result tags, output the matched standard optimization adjustment strategy, and output the corresponding energy-saving suggestion.

[0092] S4. Split the multi-source data of the air conditioning device using a random splitting method, encrypt the split data using a multi-layer encryption algorithm to obtain energy-saving data ciphertext, store different energy-saving data ciphertext in different addresses, simplify the identification of the multi-source data of the air conditioning device, obtain the simplified identification of different data of the air conditioning device, associate all simplified identifications of the air conditioning device with energy-saving values, obtain air conditioning energy-saving monitoring data, upload the air conditioning energy-saving monitoring data to the blockchain, verify the air conditioning energy-saving monitoring data through the energy-saving smart contract of the blockchain, and store the verified air conditioning energy-saving monitoring data in the blockchain;

[0093] Split the multi-source data of the air conditioning device, randomly select part of the data of the multi-source data, and the amount of the part of the data is obtained by a random function. The upper limit of the part of the data is less than half of the data amount;

[0094] The energy-saving smart contract includes energy-saving data owner verification, data submission time verification, and energy-saving location verification;

[0095] The data owner verification is verified by comparing the IP identity information of the submitter and the identity information filled in by the submitter;

[0096] The data submission time verification is verified by the maximum value and the average value of the air conditioning energy-saving data within the submission period;

[0097] The energy-saving location verification is verified by the IP address information of the submitter and the address information filled in by the submitter;

[0098] The specific steps of data encryption and decryption in step S4 are as follows:

[0099] Data encryption:

[0100] C1. Encrypt the original data using AES key A to obtain ciphertext C;

[0101] C2. Encrypt the AES key A using the RSA public key to obtain the AES key hash value h of fixed length;

[0102] C3. Combine the ciphertext C and the AES key hash value h to form the ciphertext file T;

[0103] Data decryption:

[0104] C4. Extract the ciphertext C and the AES key hash value h from the ciphertext file T;

[0105] C5. Decrypt the AES key hash value h using the RSA private key to obtain the AES key A;

[0106] C6. Decrypt the ciphertext C using the AES key A to obtain the original data.

[0107] The specific steps for obtaining the RSA public key and the RSA private key are as follows: the system randomly generates two unequal prime numbers p and q; calculate the Euler function of p and q as the RSA public key parameter n, calculate the RSA public key parameter e by the formula gcd(e,𝜑(n))=1, and take (e,n) as the RSA public key; calculate the value of d of the RSA private key using the Euclid algorithm, refer to the formula d∗e=𝑚𝑜d[(p−1)(q−1)], and take (d,n) as the RSA private key;

[0108] S5. Randomly obtain 2 energy-saving data ciphertexts of the energy-saving value of 1 air conditioning device, decrypt to obtain plaintext, and compare the plaintext with the simplified identifier in the air conditioning energy-saving monitoring data, compare the plaintext with the plaintext, and verify the data;

[0109] If the results are consistent, it means that the data is safe and has not been tampered with. If the ciphertext and the simplified identifier in the air conditioning energy-saving monitoring data are inconsistent, it means that the data has been tampered with, and the tampered position can be obtained by comparing the plaintext with the plaintext.

[0110] Embodiment 2, an air conditioning energy-saving data calculation and transmission system based on association verification, as shown in Figure 1 , comprising:

[0111] A multi-source data acquisition module for acquiring multi-source data of an air conditioning device, including sensor data of an air conditioning outdoor unit, sensor data of an air conditioning indoor unit, and multiple temperature sensor data of a room using an air conditioner;

[0112] The data verification module is configured to send the multi-source data of the air conditioning equipment into a data verification model for correlation verification and time sequence change relation verification, obtain an energy saving confidence of the multi-source data of the air conditioning equipment, and filter the multi-source data of the air conditioning equipment according to the energy saving confidence to obtain credible energy saving data, calculate an energy saving value of the air conditioning equipment by using the credible energy saving data, and remove data as energy saving abnormal data in the filtering process and mark and record the energy saving abnormal data.

[0113] The air conditioning energy saving optimization module is configured to send the credible energy saving data into an energy saving optimization model for optimization identification, obtain an optimization identification result, and make an energy saving suggestion and adjust an energy saving optimization strategy of the air conditioning equipment by using the optimization identification result.

[0114] The blockchain storage module is configured to split the multi-source data of the air conditioning equipment by using a random splitting method, encrypt the split data by using a multi-layer encryption algorithm to obtain energy saving data ciphertext, store different energy saving data ciphertext in different addresses, simplify the identification of the multi-source data of the air conditioning equipment to obtain simplified identifications of different data of the air conditioning equipment, associate all the simplified identifications of the air conditioning equipment with the energy saving value to obtain air conditioning energy saving monitoring data, upload the air conditioning energy saving monitoring data to a blockchain, verify the air conditioning energy saving monitoring data by using an energy saving smart contract of the blockchain, and store the verified air conditioning energy saving monitoring data in the blockchain.

[0115] The storage verification module is configured to randomly obtain two energy saving data ciphertexts of an energy saving value of an air conditioning equipment, decrypt the ciphertexts to obtain plaintexts, compare the plaintexts with the simplified identifications in the air conditioning energy saving monitoring data, compare the plaintexts with each other, and verify the data.

[0116] It should be understood that, for those skilled in the art, improvements or changes can be made according to the above description, and all these improvements and changes shall belong to the protection scope of the appended claims of the present application. The parts not described in detail in the specification belong to the prior art known to those skilled in the art.

Claims

1. An air conditioning energy-saving data calculation and transmission method based on association verification, characterized in that, Comprise: S1. Collecting multi-source data of air conditioning equipment, including sensor data of air conditioning outdoor unit, sensor data of air conditioning indoor unit and multiple temperature sensor data of indoor using air conditioning, obtaining a series of time series data; S2. The multi-source data of air conditioning equipment is sent to the data verification model for correlation verification and time series change relationship verification, the time series change relationship confidence and the correlation relationship confidence of the to-be-processed energy-saving data are obtained, and the energy-saving confidence of the multi-source data of the air conditioning equipment is obtained after weighted calculation, the multi-source data of the air conditioning equipment is screened according to the energy-saving confidence, the credible energy-saving data is obtained, the energy-saving value of the air conditioning equipment is calculated using the credible energy-saving data, the data eliminated in the screening process is used as energy-saving abnormal data, and the energy-saving abnormal data is marked and recorded; The calculation formula of energy-saving value is: ； D j represents the jth energy saving data, C j represents the energy saving confidence of the jth energy saving data, M j represents the maximum confidence value of the jth energy saving data, M li represents the ith rejected energy saving data, and γ represents the weight of the rejected data; S3. The credible energy-saving data is sent to the energy-saving optimization model for optimization identification, and the optimization identification result is obtained, and the air conditioning equipment is suggested and the energy-saving optimization strategy is adjusted through the optimization identification result; S4. The multi-source data of the air conditioning equipment is split using a random splitting method, and the split data is encrypted using a multi-layer encryption algorithm to obtain energy-saving data ciphertext, different energy-saving data ciphertext is stored in different addresses, the multi-source data of the air conditioning equipment is simplified and identified to obtain different data of the air conditioning equipment, and the energy-saving value is associated with all simplified identifications of the air conditioning equipment to obtain air conditioning energy-saving monitoring data, and the air conditioning energy-saving monitoring data is uploaded to the block chain, and the air conditioning energy-saving monitoring data is verified through the energy-saving smart contract of the block chain, and the verified air conditioning energy-saving monitoring data is stored in the block chain. S5. Randomly obtain two energy-saving data ciphertexts of an air conditioning equipment, decrypt the plaintext, compare the plaintext with the simplified identification in the air conditioning energy-saving monitoring data, and compare the data.

2. The method of claim 1, wherein, The data verification model comprises a data processing layer, a time series change relationship verification layer, a correlation relationship verification layer and a confidence obtaining layer; The data processing layer is used for pre-processing the obtained multi-source data to obtain to-be-processed energy-saving data; The time series relationship verification layer is used for verifying the time series change relationship of the to-be-processed energy-saving data, obtaining the time series change relationship confidence of the to-be-processed energy-saving data through the mode difference of the data in the time series change behavior, and generating the time series feature data of the to-be-processed energy-saving data; The correlation relationship verification layer is used for verifying the correlation relationship between the time series feature data of different source data, obtaining the correlation relationship confidence of the to-be-processed energy-saving data through the difference between the correlation relationship between the data and the standard correlation relationship; The confidence obtaining layer is used for weighted calculation of the time series change relationship confidence and the correlation relationship confidence, obtaining the energy-saving confidence of each data in the multi-source data, and the calculation formula of the energy-saving confidence is: ； In the formula respectively represent the time sequence change relationship confidence and the association relationship confidence, respectively are the verification weights of the time sequence change relationship confidence and the association relationship confidence, α a + α b = 1, 0.6 >= α a >= 0.2, 0.8 >= α b >= 0.

2.

3. The method of claim 2, wherein the method further comprises, Further comprising constructing a data verification model, the specific steps are: A1. Obtain multi-source data of a plurality of air conditioning devices, pre-process the multi-source data, and label time sequence change relationship confidence labels and correlation relationship confidence, obtain time sequence feature data of the pre-processed multi-source data, and group the pre-processed multi-source data and the time sequence feature data into a data verification dataset, and divide the data verification dataset into a data verification training set and a data verification test set; A2. The data verification training set is sent to the initial time sequence verification model and the initial relationship verification model based on the deep neural network to train, respectively taking the time sequence change relationship confidence label and the correlation relationship confidence as the target, to obtain the optimized time sequence verification model and the optimized relationship verification model; A3. The output verification test set is sent to the optimized time sequence verification model and the optimized relationship verification model for accuracy evaluation, respectively taking the time sequence change relationship confidence label and the correlation relationship confidence as the target, adjusting the hyperparameters of the optimized time sequence verification model and the optimized relationship verification model until the accuracy is within the error range, and the finally obtained optimized time sequence verification model and optimized relationship verification model are used as the time sequence change relationship verification layer and the correlation relationship verification layer of the data verification model; A4. Obtain the verification weights of the time sequence change relationship confidence label and the correlation relationship confidence label through the time sequence features and data relationship quantities of the multi-source data, and perform weighted calculation on the time sequence change relationship confidence and the correlation relationship confidence according to the verification weights to obtain the energy saving confidence of each data in the multi-source data.

4. The method of claim 1, wherein the method further comprises, The energy saving optimization model includes a data processing layer, an optimization identification layer and an optimization suggestion layer; The data processing layer is used for pre-processing the trusted energy saving data to obtain the to-be-identified energy saving data; The optimization identification layer is used for identifying the to-be-identified energy saving data to obtain an optimization identification result; The optimization suggestion layer is used for matching the optimization adjustment strategy in the optimization strategy library according to the optimization identification result, and generating a corresponding energy saving suggestion.

5. The method of claim 4, wherein the method further comprises, Further comprising constructing an energy saving optimization model, the specific steps are: B1. Obtain multi-source data of a plurality of air conditioning devices, pre-process the multi-source data, and label optimization identification result labels, group the pre-processed multi-source data into an energy saving optimization dataset, and divide the energy saving optimization dataset into an energy saving optimization training set and an energy saving optimization test set; B2. The energy saving optimization test set is sent to the initial energy saving optimization model based on the SVM model for training, taking the optimization identification result label as the target, to obtain a strengthened energy saving optimization model; B3. The energy saving optimization test set is sent to the strengthened energy saving optimization model for accuracy evaluation, taking the optimization identification result label as the target, adjusting the hyperparameters of the strengthened energy saving optimization model until the accuracy is within the error range, and the finally obtained strengthened energy saving optimization model is used as the optimization identification layer of the energy saving optimization model; B4. Match the standard optimization adjustment strategy in the optimization strategy library through the optimization identification result label, take the matched standard optimization adjustment strategy as the output, and output the corresponding energy saving suggestion.

6. The method of claim 1, wherein the method further comprises: Further comprising splitting the multi-source data of the air conditioning device, randomly selecting part of the multi-source data, and the amount of the part of the data is obtained by a random function, and the upper limit of the part of the data is less than half of the data amount.

7. The method of claim 1, wherein the method further comprises: Further comprising encrypting the split data using a multi-layer encryption algorithm to obtain energy-saving data ciphertext, and the specific steps of encryption and decryption are as follows: Data encryption: C1. Encrypt the original data using AES key A to obtain ciphertext C; C2. Encrypt the AES key A using the RSA public key to obtain the AES key hash value h of fixed length; C3. Combine the ciphertext C and the AES key hash value h to form the ciphertext file T; Data decryption: C4. Extract the ciphertext C and the AES key hash value h from the ciphertext file T; C5. Decrypt the AES key hash value h using the RSA private key to obtain the AES key A; C6. Decrypt the ciphertext C using the AES key A to obtain the original data.

8. The method of claim 1, wherein the method further comprises: The energy-saving smart contract includes energy-saving data owner verification, data submission time verification, and energy-saving location verification.

9. An air conditioning energy saving data calculation and transmission system based on association verification, characterized by The system applies the air conditioner energy-saving data calculation and transmission method based on association verification according to any one of claims 1-8, comprising: A multi-source data acquisition module for acquiring multi-source data of an air conditioning device, including sensor data of an air conditioner outdoor unit, sensor data of an air conditioner indoor unit, and multiple temperature sensor data of a room using an air conditioner; A data verification module for sending the multi-source data of the air conditioning device into a data verification model for association relationship verification and time sequence change relationship verification to obtain the energy-saving confidence of the multi-source data of the air conditioning device, and filtering the multi-source data of the air conditioning device according to the energy-saving confidence to obtain trusted energy-saving data, using the trusted energy-saving data to calculate the energy-saving value of the air conditioning device, and discarding the data as energy-saving abnormal data during the filtering process, and marking and recording the energy-saving abnormal data; An air conditioner energy-saving optimization module for sending the trusted energy-saving data into an energy-saving optimization model for optimization identification to obtain an optimization identification result, and making energy-saving suggestions and adjusting energy-saving optimization strategies for the air conditioning device through the optimization identification result; A blockchain storage module for splitting the multi-source data of the air conditioning device using a random splitting method, encrypting the split data using a multi-layer encryption algorithm to obtain energy-saving data ciphertext, storing different energy-saving data ciphertext in different addresses, simplifying the identification of the multi-source data of the air conditioning device to obtain different data simplification identifications of the air conditioning device, associating all the simplification identifications of the air conditioning device with the energy-saving value to obtain air conditioner energy-saving monitoring data, uploading the air conditioner energy-saving monitoring data to the blockchain, verifying the air conditioner energy-saving monitoring data through the energy-saving smart contract of the blockchain, and storing the verified air conditioner energy-saving monitoring data in the blockchain; A storage verification module for randomly obtaining two energy-saving data ciphertexts of an energy-saving value of an air conditioning device, decrypting the plaintext, comparing the plaintext with the simplification identification in the air conditioner energy-saving monitoring data, and verifying the data.

Images