A method and device for predicting oxygen content in flue gas
By generating a hash table and determining weights, and establishing a regression model, the problem of low safety of boiler flue gas oxygen content monitoring data is solved, and accurate flue gas oxygen content prediction and data safety guarantee are achieved.
Patent Information
- Application Number
- CN202011012356.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-09-23
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2040-09-23
AI Technical Summary
In the prior art, the monitoring of boiler flue gas oxygen content has the problem of low data security, especially in the complex furnace environment, the zirconia measurement points are prone to wear, and a simple hardware measurement system cannot detect the oxygen content of the flue gas. operational change characteristics of quantity.
By obtaining the characteristic data of the source and target devices, generating hash tables, calculating data distribution, determining weights, establishing a regression model, establishing a soft measurement model for predicting the oxygen content of flue gas, and achieving prediction of the oxygen content of flue gas without requiring Share the device's characteristic data to ensure data security.
Accurate prediction of the oxygen content of flue gas is achieved, data security is ensured, and security risks brought about by data sharing between multiple boilers.
Smart Images

Figure CN114254791B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of energy technology, and in particular to a method and device for predicting oxygen content in flue gas. Background Art
[0002] Flue gas oxygen content is one of the important indicators to measure whether the boiler is operating safely, economically and environmentally friendly. Zirconia is usually used in industrial production to monitor the oxygen content of flue gas at the furnace outlet. However, the environment inside the furnace is extremely complex, and the zirconia measuring points often fail to work properly due to wear. At the same time, a simple hardware measurement system cannot detect the operating change characteristics of the oxygen content of the flue gas at the furnace outlet. Therefore, it is necessary to find a practical, reliable, accurate and economical method to measure oxygen content.
[0003] At present, for the prediction of the boiler to be predicted without a label, the mapping relationship between the characteristic data of the boiler to be predicted and the flue gas oxygen content is established through data sharing between the boiler to be predicted and other boilers, and the flue gas oxygen content prediction model is obtained. The flue gas oxygen content prediction model is used to realize the prediction of the flue gas oxygen content of the boiler to be predicted.
[0004] However, the above technical solution requires sharing data among several boilers, which results in low data security among the several boilers. Summary of the invention
[0005] In view of the above-mentioned technical problems existing in the prior art, the present invention provides a method, device, computer-readable storage medium and electronic device for predicting the oxygen content of flue gas, and the characteristic data of the boiler does not need to be shared between the source device and the target device, thereby ensuring data security.
[0006] In a first aspect, the present invention provides a method for predicting oxygen content in flue gas, comprising:
[0007] Acquire characteristic data of a source device and characteristic data of a target device, wherein the characteristic data of the source device is non-shared data;
[0008] Calculating data distribution of the characteristic data of the source device relative to the target device based on a hash table generated from the characteristic data of the target device;
[0009] Determining a weight of the feature data of the source device relative to the target device according to a data distribution of the feature data of the source device relative to the target device;
[0010] According to the weight of the characteristic data of the source device relative to the target device and the historical smoke content corresponding to the characteristic data of the source device, a relationship between the regression model and the smoke oxygen content is established to obtain a smoke oxygen content prediction soft measurement model;
[0011] The flue gas oxygen content of the target device is predicted according to the flue gas oxygen content prediction soft measurement model.
[0012] In a second aspect, the present invention provides a method for predicting oxygen content in flue gas, comprising:
[0013] A data acquisition module, used to acquire characteristic data of a source device and characteristic data of a target device, wherein the characteristic data of the source device is non-shared data;
[0014] A distribution calculation module, configured to calculate data distribution of the characteristic data of the source device relative to the target device based on a hash table generated by the characteristic data of the target device;
[0015] A weight determination module, configured to determine a weight of the feature data of the source device relative to the target device according to a data distribution of the feature data of the source device relative to the target device;
[0016] A model determination module, used to establish a relationship between the regression model and the smoke oxygen content according to the weight of the characteristic data of the source device relative to the target device and the historical smoke content corresponding to the characteristic data of the source device, so as to obtain a smoke oxygen content prediction soft measurement model;
[0017] The prediction module is used to predict the flue gas oxygen content of the target device according to the flue gas oxygen content prediction soft measurement model.
[0018] In a third aspect, the present invention provides a computer-readable storage medium, comprising execution instructions. When a processor of an electronic device executes the execution instructions, the processor executes any method described in the first aspect.
[0019] In a fourth aspect, the present invention provides an electronic device, comprising a processor and a memory storing execution instructions, wherein when the processor executes the execution instructions stored in the memory, the processor executes any method described in the first aspect.
[0020] The present invention provides a method, device, computer-readable storage medium and electronic device for predicting flue gas oxygen content. The method obtains characteristic data of a source device and characteristic data of a target device, wherein the characteristic data of the source device is non-shared data, and then, based on a hash table generated from the characteristic data of the target device, calculates the data distribution of the characteristic data of the source device relative to the target device, and then, based on the data distribution of the characteristic data of the source device relative to the target device, determines the weight of the characteristic data of the source device relative to the target device, and then, based on the weight of the characteristic data of the source device relative to the target device and the historical flue gas content corresponding to the characteristic data of the source device, establishes a relationship between a regression model and flue gas oxygen content, obtains a flue gas oxygen content prediction soft measurement model, and then, based on the flue gas oxygen content prediction soft measurement model, predicts the flue gas oxygen content of the target device. The technical solution provided by the present invention establishes a data association between the source device and the target device through a hash table of the non-shared characteristic data of the source device and the characteristic data of the target device, migrates the data of the source device to the target device, and does not require the characteristic data of the device to be shared between the source device and the target device, thereby ensuring data security.
[0021] The further effects of the above-mentioned non-conventional preferred manner will be described below in conjunction with specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] In order to more clearly illustrate the embodiments of the present invention or the existing technical solutions, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0023] Figure 1 A schematic flow chart of a method for predicting oxygen content in flue gas provided by one embodiment of the present invention;
[0024] Figure 2 A schematic flow chart of another method for predicting oxygen content in flue gas provided by one embodiment of the present invention;
[0025] Figure 3 A schematic diagram of the structure of a device for predicting oxygen content in flue gas provided by one embodiment of the present invention;
[0026] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0027] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. 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.
[0028] like Figure 1 As shown, an embodiment of the present invention provides a method for predicting flue gas oxygen content, which is applied to an edge node of a boiler to be predicted, and includes the following steps:
[0029] Step 101: Acquire feature data of a source device and feature data of a target device, wherein the feature data of the source device is non-shared data.
[0030] Specifically, the source device and the target device are both devices that can generate flue gas, such as gas boilers. The target device and the source device have the same device type, such as gas boilers, but the production unit, device model and other information may be different.
[0031] Specifically, there are multiple feature data of the target device, each of which includes several feature values corresponding to the target device, and several features are factors affecting the oxygen content of flue gas of the target device, which need to be determined in combination with the actual scenario. For example, if the target device is a gas boiler, the several features include but are not limited to gas flow, gas temperature, exhaust temperature, flue gas flow, gas pressure, flue gas humidity and flue gas pressure, etc. There are multiple feature data of the source device, each of which includes several feature values corresponding to the source device, wherein several features are the same as the features corresponding to the feature data of the target device.
[0032] It should be noted that the characteristic data of the source device is non-shared data, thereby ensuring the data security of the characteristic data of the source device.
[0033] Optionally, multiple feature data of the target device are obtained through the edge node corresponding to the target device. The edge node of the target device can be understood as a node that is closest to the target device and can perform data processing and data interaction. The node includes but is not limited to any one or more of an edge server, an edge gateway, and an edge controller.
[0034] Optionally, multiple feature data of the source device are obtained through the edge node corresponding to the source device. The edge node of the source device can be understood as the node closest to the source device and capable of data processing and data interaction. The node includes but is not limited to any one or more of an edge server, an edge gateway, and an edge controller. Correspondingly, if the feature data of the source device is non-shared data, it means that the feature data of the source device does not come from the edge node corresponding to the source device.
[0035] Step 102: Calculate data distribution of the characteristic data of the source device relative to the target device based on a hash table generated from the characteristic data of the target device.
[0036] In this embodiment, a hash table generated by the characteristic data of the target device is used to calculate the data distribution of the characteristic data of the source device relative to the target device, thereby establishing a connection between the characteristic data of the origin device and the characteristic data of the target device.
[0037] Specifically, for each feature data of the target device, the feature data is converted through a hash function to obtain a hash value of the feature data; and a hash table is obtained through the hash values of all feature data of the target device. The hash value can be understood as the value obtained after the feature data of the target device is keylessly encrypted through a hash function. It should be noted that, considering that the hash function can realize keyless encryption processing of data relatively quickly and simply, the embodiment of the present invention selects a hash function to perform keyless encryption on the feature data of the target device.
[0038] Specifically, the data distribution may be a distribution probability or a distribution frequency, and preferably a distribution probability.
[0039] In one embodiment, the data distribution of the characteristic data of the source device relative to the data of the target device may be calculated in the following manner:
[0040] Obtain a hash value of the characteristic data of the source device; determine the distribution probability of the characteristic data of the source device relative to the target device based on the hash table and hash value generated by the characteristic data of the target device, and determine the distribution probability as the data distribution of the characteristic data of the source device relative to the target device.
[0041] In this embodiment, data distribution is obtained based on hash values and hash tables, and does not involve comparison between feature data. There is no need to share feature data between source devices and target devices, thus ensuring data security.
[0042] Specifically, the hash value of the characteristic data of the source device can be understood as the value obtained after keyless encryption of the characteristic data of the source device. There is no need to share the characteristic data of the source device. Only the hash value of the characteristic data of the source device needs to be used for information sharing, thereby ensuring data security between the source device and the target device. It should be noted that when the parameters in the hash function are different, the difficulty of analyzing the hash value and the hash table may increase. Therefore, the hash value and the hash table are obtained by keyless encryption based on the same hash function, thereby reducing the difficulty of analyzing the hash value and the hash table.
[0043] Optionally, the hash value of the characteristic data of the source device is obtained by the edge node corresponding to the source device in the Internet of Things. The node can perform keyless encryption on the characteristic data of the source device based on the hash function that generates the hash table to obtain the hash value of the characteristic data of the source device.
[0044] Optionally, based on the hash table and hash value generated by the characteristic data of the target device, the replacement data of the characteristic data of the source device is determined from the characteristic data of the target device, and the distribution probability of the replacement data in the characteristic data of the target device is determined as the distribution probability of the characteristic data of the source device relative to the target device. Here, through the hash value and the hash table, the replacement data corresponding to the characteristic data of the source device is simply and quickly determined from the characteristic data of the target device, thereby establishing a data association between the target device and the source device. As a possible implementation method, based on the local sensitive hashing algorithm, the hash table and the hash value of the characteristic data of the source device are analyzed for similarity to determine the replacement data of the characteristic data of the source device from each characteristic data of the target device. Here, the local sensitive hashing algorithm can save computing time and improve computing efficiency. It should be noted that when multiple replacement data of the characteristic data of the source device are determined from each characteristic data of the target device, one can be randomly selected as the replacement data.
[0045] Here, considering that the non-parametric estimation method does not need to rely on any prior assumptions about the data, but is determined based on the data itself, and is simpler than the parameter estimation method, the embodiment of the present invention selects the non-parametric estimation method to determine the distribution probability. It should be further explained that the present invention does not intend to limit the non-parametric estimation method, which can be determined in combination with actual conditions. Considering that the amount of feature data in the embodiment of the present invention is relatively large, in order to more accurately estimate the data distribution, the kernel density estimation method is preferred.
[0046] Optionally, based on a kernel density estimation method, the distribution probability of the replacement data in the feature data of the target device is determined.
[0047] Step 103: Determine the weight of the feature data of the source device relative to the target device according to the data distribution of the feature data of the source device relative to the target device.
[0048] In this embodiment, by determining the weight of the characteristic data of the source device relative to the target device, a data association is established between the target device and the source device, thereby achieving migration of the data of the source device to the target device without sharing the characteristic data between the source device and the target device.
[0049] In one embodiment, the weight of the characteristic data of the source device relative to the target device may be determined in the following manner:
[0050] Determine the distribution probability of the characteristic data of the source device; determine the weight of the characteristic data of the source device relative to the target device as the ratio of the distribution probability of the characteristic data of the source device relative to the target device and the distribution probability of the characteristic data of the source device.
[0051] In this embodiment, the weight of the characteristic data of the source device relative to the target device is determined by the ratio of the distribution probability of the characteristic data of the source device relative to the target device and the distribution probability of the characteristic data of the source device, so that the characteristic data of the source device is migrated to the target device, thereby realizing data migration between the target device and the source device without the need for data sharing, thereby ensuring data security.
[0052] Specifically, the edge node of the source device sends the hash value of the feature data of the source device to the edge node of the target device. The edge node of the target device determines the distribution probability of the feature data of the source device relative to the target device based on the hash table and the hash value of the feature data of the source device, and sends it to the edge node of the source device. The edge node of the source device determines, for each feature data of the source device, the ratio of the distribution probability of the feature data relative to the target device and the distribution probability relative to each feature data of the source device, and determines the ratio as the weight of the feature data of the source device relative to the target device.
[0053] Optionally, based on a non-parametric estimation method, the distribution probability of the characteristic data of the source device is determined, and the non-parametric estimation method is preferably a kernel density estimation method. Specifically, based on the non-parametric estimation method, a probability density model of the characteristic data of the source device is determined; the characteristic data of the source device is substituted into the probability density model to determine the distribution probability of the characteristic data of the source device.
[0054] It should be noted that the edge node corresponding to the source device sends the hash value of the source device's feature data, while the edge node corresponding to the target device sends the distribution probability of the source device's feature data relative to the target device. There is no data sharing involved between the target device and the source device, thus ensuring data security.
[0055] Step 104: According to the weight of the characteristic data of the source device relative to the target device and the historical flue gas content corresponding to the characteristic data of the source device, a relationship between the regression model and the flue gas oxygen content is established to obtain a flue gas oxygen content prediction soft measurement model.
[0056] In this embodiment, the relationship between the regression model and the flue gas oxygen content is established through the weight of the characteristic data of the source device relative to the target device and the historical flue gas content corresponding to the characteristic data of the source device, and a flue gas oxygen content prediction soft measurement model is obtained. There is no need to share the characteristic data of the source device, thereby ensuring data security.
[0057] In one embodiment, the flue gas oxygen content prediction soft sensor model can be determined by the following method:
[0058] The regression model is trained according to the weight of the characteristic data of the source device relative to the target device and the historical flue gas content corresponding to the characteristic data of the source device, and the trained regression model is determined as a flue gas oxygen content prediction soft measurement model, which indicates the relationship between the characteristic data of the target device and the flue gas oxygen content.
[0059] Specifically, the edge node corresponding to the source device trains a regression model based on the weights of multiple feature data of the source device relative to the target device and the corresponding historical flue gas oxygen content, and determines the trained regression model as a flue gas oxygen content prediction soft measurement model. The trained flue gas oxygen content prediction soft measurement model takes into account the weight of the feature data of the source device relative to the target device, and therefore can indicate the relationship between the feature data of the target device and the flue gas oxygen content.
[0060] Specifically, the characteristic data of the source device and the weight of the characteristic data of the source device relative to the target device can realize the migration of the data of the source device to the target device. Subsequently, the parameters of the regression model are adjusted according to the weights corresponding to multiple characteristic data of the source device, so that the parameters of the adjusted regression model can reflect the relationship between the characteristic data of the target device and the oxygen content of the flue gas, and do not involve data sharing between the source device and the target device, thereby ensuring data security.
[0061] Specifically, the regression model can be any regression model in the prior art, such as a neural network model, and is not specifically limited here.
[0062] It should be noted that the data of the target device and the source device are distributed in different edge nodes in the Internet of Things. Sharing data for model training will cause data security issues. The embodiment of the present invention calculates the weight of the characteristic data of the source device based on the data distribution of the characteristic data of the two nodes in the Internet of Things, thereby ensuring that in the process of obtaining the flue gas oxygen content model, the data between the target device and the source device does not leave their edge nodes in the Internet of Things, thereby ensuring the privacy of the data.
[0063] Step 105: predict the flue gas oxygen content of the target device according to the flue gas oxygen content prediction soft measurement model.
[0064] Specifically, the current characteristic data of the target device is collected, and the current characteristic data is substituted into the prediction of the flue gas oxygen content, so that the flue gas oxygen content of the target device can be determined.
[0065] In order to better understand the data processing process between the edge node of the target device and the edge node of the source device, for example, assume that the edge node of the target device is set to A, and the edge node of the source device is set to B. B sends the hash value corresponding to each feature data of the source device to B, and then receives the distribution probability of each feature data sent by A relative to the target device, and combines the distribution probability of the feature data in each feature data of the source device to determine the weight of each feature data relative to the target device. After that, the regression model is trained using the weight corresponding to each feature data and the historical flue gas oxygen content to obtain a flue gas oxygen content prediction soft measurement model, and the flue gas oxygen content prediction soft measurement model is sent to A, and A predicts the flue gas oxygen content of the target device through the flue gas oxygen content prediction soft measurement model.
[0066] It can be seen from the above technical solutions that the embodiments of the present invention have at least the following effective effects: through the hash table of the characteristic data of the target device, a data association is established between the source device and the target device, the weight of the characteristic data of the source device relative to the target device is obtained, and the characteristic data of the source device is migrated to the target device, without sharing the characteristic data between the source device and the target device, thereby ensuring data security.
[0067] Figure 1 What is shown is only a basic embodiment of the method of the present invention. By performing certain optimization and expansion on this basis, other preferred embodiments of the method can be obtained.
[0068] In order to more clearly illustrate the technical solution of the present invention, please refer to Figure 2 The embodiment of the present invention provides another method for predicting the oxygen content of flue gas. This embodiment is based on the above embodiment and is further described in combination with a specific application scenario. In this embodiment, the following steps may be specifically included:
[0069] Step 201: Acquire feature data of a source device and feature data of a target device, wherein the feature data of the source device is non-shared data.
[0070] Assume that the edge node corresponding to the target device is A, and the edge node corresponding to the source device is B. A can obtain all feature data of the target device, and B can obtain all feature data of the source device.
[0071] Step 202: Based on the local sensitive hashing algorithm, a similarity analysis is performed on the hash table generated according to the feature data of the target device and the hash value of the feature data of the source device to determine the replacement data of the feature data of the source device from the feature data of the target device.
[0072] A performs keyless encryption on each characteristic data of the target device according to the hash function to obtain a hash table, and sends the hash function to B.
[0073] B accepts the hash function sent by A, and performs keyless encryption on the characteristic data of the source device according to the hash function to obtain the hash value of the characteristic data of the source device, and sends it to A.
[0074] A receives the hash value of the characteristic data of the source device sent by B, and based on the local sensitive hashing algorithm, performs a similarity analysis between the hash table and the hash value of the characteristic data of the source device to determine the replacement data of the characteristic data of the target device from the characteristic data of the source device.
[0075] Step 203: Determine the distribution probability of the replacement data in the characteristic data of the target device as the distribution probability of the characteristic data of the source device relative to the target device.
[0076] A sends the distribution probability of the source device's feature data relative to the target device to B.
[0077] Step 204: Determine a probability density model of the characteristic data of the source device based on a non-parametric estimation method; and determine the distribution probability of the characteristic data of the source device according to the probability density model.
[0078] B calculates the distribution probability of the feature data in the feature data of the source device based on the kernel density estimation algorithm.
[0079] Step 205: Determine the ratio of the distribution probability of the characteristic data of the source device relative to the target device to the distribution probability of the characteristic data of the source device as the weight of the characteristic data of the source device relative to the target device.
[0080] B calculates the ratio of the distribution probability of the feature data of the source device relative to the target device and the distribution probability of the feature data relative to the source device, and determines the ratio as the weight of the source device relative to the target device.
[0081] Step 206: Train a regression model based on the weight of the characteristic data of the source device relative to the target device and the historical flue gas content corresponding to the characteristic data of the source device, and determine the trained regression model as a flue gas oxygen content prediction soft measurement model, wherein the flue gas oxygen content prediction soft measurement model indicates the relationship between the characteristic data of the target device and the flue gas oxygen content.
[0082] B trains a regression model based on the weight of the characteristic data of the source device relative to the target device and the historical flue gas content corresponding to the characteristic data of the source device, and sends the trained regression model to A as a soft measurement model for flue gas oxygen content prediction.
[0083] Step 207: predict the flue gas oxygen content of the target device according to the flue gas oxygen content prediction soft measurement model.
[0084] A receives the flue gas oxygen content prediction soft measurement model sent by B, and uses the flue gas oxygen content prediction soft measurement model to predict the flue gas oxygen content of the target equipment.
[0085] It can be seen from the above technical solution that the beneficial effect of this embodiment is: according to the data distribution of the characteristic data of two nodes in the Internet of Things, the weight of the characteristic data of the source device is calculated based on the hash value and the hash table, so that the characteristic data of the source device is migrated to the target device, and the data between the target device and the source device does not leave its edge node in the Internet of Things, thereby ensuring the privacy of the data.
[0086] Based on the same concept as the flue gas oxygen content prediction method provided by the method embodiment of the present invention, please refer to Figure 3 The present invention implements a flue gas oxygen content prediction device, which is applied to the edge node corresponding to the boiler to be predicted, including:
[0087] A data acquisition module 301 is used to acquire characteristic data of a source device and characteristic data of a target device, wherein the characteristic data of the source device is non-shared data;
[0088] A distribution calculation module 302, configured to calculate data distribution of the characteristic data of the source device relative to the target device according to a hash table generated by the characteristic data of the target device;
[0089] A weight determination module 303, configured to determine a weight of the feature data of the source device relative to the target device according to data distribution of the feature data of the source device relative to the target device;
[0090] A model determination module 304 is used to establish a relationship between a regression model and smoke oxygen content according to the weight of the characteristic data of the source device relative to the target device and the historical smoke content corresponding to the characteristic data of the source device, so as to obtain a smoke oxygen content prediction soft measurement model;
[0091] The prediction module 305 is used to predict the flue gas oxygen content of the target device according to the flue gas oxygen content prediction soft measurement model.
[0092] In one embodiment of the present invention, the distribution calculation module 302 includes: a hash value calculation unit and a distribution calculation unit; wherein,
[0093] The hash value calculation unit is used to calculate the hash value of the characteristic data of the source device;
[0094] The distribution calculation unit is used to determine the distribution probability of the characteristic data of the source device relative to the target device based on the hash table generated by the characteristic data of the target device and the hash value, and determine the distribution probability as the data distribution of the characteristic data of the source device relative to the target device.
[0095] In one embodiment of the present invention, the distribution calculation unit includes: a retrieval subunit and a probability determination subunit; wherein,
[0096] The retrieval subunit is used to determine the replacement data of the characteristic data of the source device from the characteristic data of the target device according to the hash table generated by the characteristic data of the target device and the hash value;
[0097] The probability determination subunit is used to determine the distribution probability of the replacement data in the characteristic data of the target device as the distribution probability of the characteristic data of the source device relative to the target device.
[0098] In one embodiment of the present invention, the retrieval subunit is used to perform a similarity analysis on a hash table generated according to the feature data of the target device and a hash value of the feature data of the source device based on a local sensitive hashing algorithm, so as to determine replacement data for the feature data of the source device from the feature data of the target device.
[0099] In one embodiment of the present invention, the weight determination module 302 includes: a probability determination unit and a weight determination unit; wherein,
[0100] The probability determination unit is used to determine the distribution probability of the characteristic data of the source device;
[0101] The weight determination unit is used to determine the ratio of the distribution probability of the feature data of the source device relative to the target device and the distribution probability of the feature data of the source device as the weight of the feature data of the source device relative to the target device.
[0102] In one embodiment of the present invention, the probability determination unit is used to determine a probability density model of the characteristic data of the source device based on a non-parametric estimation method; and determine the distribution probability of the characteristic data of the source device according to the probability density model.
[0103] In one embodiment of the present invention, the model determination module 304 is used to train a regression model based on the weight of the characteristic data of the source device relative to the target device and the historical flue gas content corresponding to the characteristic data of the source device, and determine the trained regression model as a flue gas oxygen content prediction soft measurement model, wherein the flue gas oxygen content prediction soft measurement model indicates the relationship between the characteristic data of the target device and the flue gas oxygen content.
[0104] Figure 4 It is a structural diagram of an electronic device provided by an embodiment of the present invention. At the hardware level, the electronic device includes a processor 401 and a memory 402 storing execution instructions, and optionally also includes an internal bus 403 and a network interface 404. Among them, the memory 402 may include a memory 4021, such as a high-speed random access memory (Random-Access Memory, RAM), and may also include a non-volatile memory 4022 (non-volatile memory), such as at least one disk storage, etc.; the processor 401, the network interface 404 and the memory 402 can be interconnected through an internal bus 403, and the internal bus 403 can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc.; the internal bus 403 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one bidirectional arrow is used in the figure, but it does not mean that there is only one bus or one type of bus. Of course, the electronic device may also include hardware required for other services. When the processor 401 executes the execution instruction stored in the memory 402, the processor 401 executes the method in any embodiment of the present invention, and is at least used to execute the following steps: Figure 1 or Figure 2 The method shown.
[0105] In a possible implementation, the processor reads the corresponding execution instruction from the non-volatile memory into the memory and then runs it, and can also obtain the corresponding execution instruction from other devices to form a flue gas oxygen content prediction device at the logical level. The processor executes the execution instruction stored in the memory to implement a flue gas oxygen content prediction method provided in any embodiment of the present invention through the execution of the execution instruction.
[0106] The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or instructions in the form of software. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, steps and logic block diagrams in the embodiments of the present invention can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0107] The embodiment of the present invention further provides a computer-readable storage medium, including an execution instruction. When a processor of an electronic device executes the execution instruction, the processor executes the method provided in any embodiment of the present invention. The electronic device may be specifically as follows: Figure 4 The electronic device shown; the execution instruction is a computer program corresponding to a method for predicting the oxygen content of flue gas.
[0108] Those skilled in the art will appreciate that the embodiments of the present invention may be provided as methods or computer program products. Therefore, the present invention may be implemented in the form of a complete hardware embodiment, a complete software embodiment, or a combination of software and hardware.
[0109] The multiple embodiments of the present invention are described in a progressive manner, and the same or similar parts between the multiple embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0110] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0111] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.
Claims
1. A method for predicting flue gas oxygen content, characterized in that: include: Acquire characteristic data of a source device and characteristic data of a target device, wherein the characteristic data of the source device is non-shared data; Calculating data distribution of the characteristic data of the source device relative to the target device based on a hash table generated from the characteristic data of the target device; Determining a weight of the feature data of the source device relative to the target device according to a data distribution of the feature data of the source device relative to the target device; According to the weight of the characteristic data of the source device relative to the target device and the historical smoke content corresponding to the characteristic data of the source device, a relationship between the regression model and the smoke oxygen content is established to obtain a smoke oxygen content prediction soft measurement model; Predicting the flue gas oxygen content of the target device according to the flue gas oxygen content prediction soft measurement model; The calculating the data distribution of the characteristic data of the source device relative to the target device by generating a hash table based on the characteristic data of the target device includes: Calculating a hash value of the characteristic data of the source device; Determine, based on the hash table generated by the characteristic data of the target device and the hash value, a distribution probability of the characteristic data of the source device relative to the target device, and determine the distribution probability as a data distribution of the characteristic data of the source device relative to the target device; The determining, according to data distribution of the feature data of the source device relative to the target device, a weight of the feature data of the source device relative to the target device, comprises: Determining a distribution probability of characteristic data of the source device; The ratio of the distribution probability of the feature data of the source device relative to the target device to the distribution probability of the feature data of the source device is determined as the weight of the feature data of the source device relative to the target device.
2. The method according to claim 1, characterized in that The determining, based on the hash table generated according to the feature data of the target device and the hash value, the distribution probability of the feature data of the source device relative to the target device comprises: Determining replacement data for the characteristic data of the source device from the characteristic data of the target device according to the hash table generated by the characteristic data of the target device and the hash value; The distribution probability of the replacement data in the characteristic data of the target device is determined as the distribution probability of the characteristic data of the source device relative to the target device.
3. The method according to claim 2, characterized in that The step of generating a hash table and a hash value based on the characteristic data of the target device and determining replacement data for the characteristic data of the source device from the characteristic data of the target device comprises: Based on the local sensitive hashing algorithm, a similarity analysis is performed on a hash table generated according to the feature data of the target device and a hash value of the feature data of the source device to determine replacement data of the feature data of the source device from the feature data of the target device.
4. The method according to claim 1, characterized in that: The determining the distribution probability of the characteristic data of the source device includes: Determine a probability density model of the characteristic data of the source device based on a non-parametric estimation method; The distribution probability of the characteristic data of the source device is determined according to the probability density model.
5. The method according to claim 1, characterized in that The method of establishing a relationship between a regression model and flue gas oxygen content according to the weight of the characteristic data of the source device relative to the target device and the historical flue gas content corresponding to the characteristic data of the source device, and obtaining a flue gas oxygen content prediction soft measurement model, includes: A regression model is trained according to the weight of the characteristic data of the source device relative to the target device and the historical flue gas content corresponding to the characteristic data of the source device, and the trained regression model is determined as a flue gas oxygen content prediction soft measurement model, which indicates the relationship between the characteristic data of the target device and the flue gas oxygen content.
6. A device for predicting flue gas oxygen content, characterized in that: include: A data acquisition module, used to acquire characteristic data of a source device and characteristic data of a target device, wherein the characteristic data of the source device is non-shared data; A distribution calculation module, configured to calculate data distribution of the characteristic data of the source device relative to the target device based on a hash table generated by the characteristic data of the target device; A weight determination module, configured to determine a weight of the feature data of the source device relative to the target device according to a data distribution of the feature data of the source device relative to the target device; A model determination module, used to establish a relationship between the regression model and the smoke oxygen content according to the weight of the characteristic data of the source device relative to the target device and the historical smoke content corresponding to the characteristic data of the source device, so as to obtain a smoke oxygen content prediction soft measurement model; A prediction module, used for predicting the flue gas oxygen content of the target device according to the flue gas oxygen content prediction soft measurement model; The distributed calculation module includes: a hash value calculation unit and a distributed calculation unit; wherein, The hash value calculation unit is used to calculate the hash value of the characteristic data of the source device; The distribution calculation unit is used to determine the distribution probability of the characteristic data of the source device relative to the target device according to the hash table generated by the characteristic data of the target device and the hash value, and determine the distribution probability as the data distribution of the characteristic data of the source device relative to the target device; The weight determination module includes: a probability determination unit and a weight determination unit; wherein, The probability determination unit is used to determine the distribution probability of the characteristic data of the source device; The weight determination unit is used to determine the ratio of the distribution probability of the feature data of the source device relative to the target device and the distribution probability of the feature data of the source device as the weight of the feature data of the source device relative to the target device.
7. A readable medium comprising an execution instruction. When a processor of an electronic device executes the execution instruction, the electronic device executes the method according to any one of claims 1 to 5.
8. An electronic device, comprising a processor and a memory storing execution instructions, wherein when the processor executes the execution instructions stored in the memory, the processor executes the method according to any one of claims 1 to 5.
