A method, apparatus and system for processing multi-sensor data

Abstract

This invention provides a multi-sensor data processing method, apparatus, and system with the following advantages: Monitoring data is grouped according to data elements to obtain multiple data groups, where each data element includes multiple physical parameters; these data groups are fitted separately, and the data with the smallest fitting error is selected from each group to form a training set; the mathematical relationship between the data elements is calculated using the training set. This invention groups sensor data for processing, uses high-precision sensor data for fitting calculations, shields against data noise interference, and effectively improves the accuracy of sensor data processing.

Description

Technical Field

[0001] This invention relates to the field of sensor data detection, and specifically to a multi-sensor data processing method, apparatus, and system. Background Technology

[0002] In modern science and engineering, sensor technology is widely used in data acquisition and measurement processes. Sensors can monitor various physical quantities, such as temperature, pressure, humidity, and light intensity, and their measurement data is of great significance for real-time monitoring, control, and decision-making. Wavelength-temperature-pressure correlation is a common and important problem in sensor data processing. However, sensor measurements are often subject to external noise and interference, leading to errors in the measurement data. Especially in wavelength-temperature-pressure correlation fitting, due to changes in the external environment and the influence of noise, wavelength data measured at the same temperature and pressure may not be unique.

[0003] Currently, several fitting algorithms exist for handling wavelength-temperature-pressure correlation problems, including polynomial fitting, least squares method, and Gaussian fitting. However, these traditional methods are easily affected by instrument measurement errors when faced with external noise and non-unique wavelengths, thus reducing fitting accuracy. Using the average of multiple measurements at a specific wavelength to fit the pressure is the most common method for handling errors caused by external noise, but this method cannot effectively improve fitting accuracy when there is limited measurement data.

[0004] Therefore, how to effectively improve the accuracy of sensor data processing and reduce errors is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] To overcome the shortcomings of existing sensor data processing methods, such as poor accuracy and large errors, this invention proposes a multi-sensor data processing method, device, and system.

[0006] To achieve the above objectives, according to a first aspect of the present invention, an embodiment of the present invention provides a multi-sensor data processing method, the method comprising the following steps:

[0007] The monitoring data is grouped according to data elements to obtain multiple data groups, and the data elements include multiple physical parameters;

[0008] The multiple data sets are fitted separately, and the data with the smallest fitting error is selected from each data set and combined to obtain the training set.

[0009] The mathematical relationships between data elements are obtained by fitting the training set.

[0010] Optionally, when the data elements include wavelength, temperature, and pressure, the monitoring data is grouped according to the data elements to obtain multiple data groups, including:

[0011] Wavelengths under the same temperature and pressure conditions are extracted from the monitoring data to obtain multiple reference groups;

[0012] Data is extracted from the multiple reference groups in chronological order to obtain multiple data groups.

[0013] Optionally, when the data elements include wavelength, temperature, and pressure, the monitoring data is grouped according to the data elements to obtain multiple data groups, including:

[0014] Wavelengths under the same temperature and pressure conditions are extracted from the monitoring data to obtain multiple reference groups;

[0015] Data is randomly extracted from the multiple reference groups to obtain multiple data groups.

[0016] Optionally, the step of extracting wavelengths from the monitoring data under conditions of identical temperature and pressure to obtain multiple reference groups includes:

[0017] The wavelengths under the same temperature and pressure conditions are grouped into multiple initial groups;

[0018] Calculate the wavelength dispersion in the initial group;

[0019] When the dispersion is greater than or equal to the first threshold, delete the data with the largest error in the initial group, and calculate the wavelength dispersion in the initial group after deletion.

[0020] When the dispersion is less than the first threshold, the initial group corresponding to the dispersion is used as the reference group.

[0021] Optionally, selecting the data with the smallest fitting error from each data set includes:

[0022] Calculate the fitting error, which includes one or more of the following: mean square error, root mean square error, and mean absolute error;

[0023] When the error is greater than or equal to the second threshold, the data with the largest error in the data group is deleted to obtain a new data group, and the new data group is fitted.

[0024] When the error is less than the second threshold, the data with the smallest fitting error is selected from the data set.

[0025] According to a second aspect of the present invention, embodiments of the present invention also provide a multi-sensor data processing apparatus, comprising:

[0026] The grouping module is used to group the monitoring data according to data elements to obtain multiple data groups, wherein the data elements include multiple physical parameters;

[0027] The training set generation module is used to fit the multiple data groups separately, select the data with the smallest fitting error from each data group, and combine them to obtain the training set.

[0028] The calculation module is used to fit and calculate the mathematical relationships between data elements using the training set.

[0029] Optionally, the grouping module is further used to,

[0030] When the data element includes wavelength, temperature and pressure, the wavelength under the condition that the temperature and pressure are the same is extracted from the monitoring data to obtain multiple reference groups;

[0031] Data is extracted from the multiple reference groups in chronological order or randomly to obtain multiple data groups.

[0032] Optionally, the grouping module is further used to,

[0033] The wavelengths under the same temperature and pressure conditions are grouped into multiple initial groups;

[0034] Calculate the wavelength dispersion in the initial group;

[0035] When the dispersion is greater than or equal to the first threshold, delete the data with the largest error in the initial group, and calculate the wavelength dispersion in the initial group after deletion.

[0036] When the dispersion is less than the first threshold, the initial group corresponding to the dispersion is used as the reference group.

[0037] Optionally, the training set generation module is further used to,

[0038] Calculate the fitting error, which includes one or more of the following: mean square error, root mean square error, and mean absolute error.

[0039] When the error is greater than or equal to the second threshold, the data with the largest error in the data group is deleted to obtain a new data group, and the new data group is fitted.

[0040] When the error is less than the second threshold, the data with the smallest fitting error is selected from the data set.

[0041] According to a third aspect of the present invention, embodiments of the present invention also provide a multi-sensor data processing system, comprising:

[0042] A laser controller is used to control a laser to emit laser light at a preset wavelength.

[0043] The data acquisition unit is used to obtain the monitoring data from the sensor after photoelectric conversion and send it to the processor;

[0044] The system includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the multi-sensor data processing method described in the above embodiments.

[0045] As described above, the multi-sensor data processing method, apparatus, and system provided by the embodiments of the present invention have at least the following beneficial effects: by grouping monitoring data according to data elements to obtain multiple data groups, fitting each data group separately, selecting the data with the smallest fitting error from each data group, combining them to obtain a training set, and using the training set to fit and calculate the mathematical relationship between data elements. The present invention groups sensor data for processing, uses high-precision sensor data for fitting calculations, shields data noise interference, and effectively improves the accuracy of sensor data processing. Attached Figure Description

[0046] Figure 1 This is a flowchart illustrating a multi-sensor data processing method provided in an embodiment of the present invention;

[0047] Figure 2 This is a flowchart illustrating a data grouping method provided in an embodiment of the present invention;

[0048] Figure 3 This is a flowchart illustrating another data grouping method provided in an embodiment of the present invention;

[0049] Figure 4 This is a flowchart illustrating a reference group data processing method provided in an embodiment of the present invention;

[0050] Figure 5 This is a flowchart illustrating the first data filtering method provided in an embodiment of the present invention;

[0051] Figure 6 An example is shown comparing the error to the average fitting error;

[0052] Figure 7 An example is shown comparing the fit with the outlier-removed average;

[0053] Figure 8 Examples are provided showing the comparison of the maximum value of the fitting error with the average under different data volumes;

[0054] Figure 9 An example is shown, comparing the maximum fitting error of different fitting methods;

[0055] Figure 10 This is a schematic diagram of the structure of a multi-sensor data processing device provided in an embodiment of the present invention;

[0056] Figure 11 This is a schematic diagram of the structure of the multi-sensor data processing system provided in an embodiment of the present invention. Detailed Implementation

[0057] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0058] Please see Figures 1 to 11 It should be noted that the illustrations provided in this embodiment are only schematic representations of the basic concept of the present invention. Therefore, the illustrations only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0059] The multi-sensor data processing method provided in this invention is used to address measurement errors caused by external noise in sensor measurement data, and is applied in the field and scenarios of sensor data monitoring.

[0060] To facilitate the explanation of the multi-sensor data processing method of this invention, in an exemplary embodiment, the multi-sensor data processing system may include a laser control module, an FPGA (Field Programmable Gate Array) acquisition module, and an algorithm module. Various fiber Bragg grating sensors and various MEMS (Micro-Electro-Mechanical System) optical sensors are externally connected. This system can monitor pressure changes in aircraft bodies or blades using pressure sensors in fields such as aerospace and wind power generation. This method can be applied to various sensor data fitting scenarios, not limited to pressure measurement using pressure sensors, but also to scenarios where sensor values ​​are affected by temperature in addition to pressure wavelength, such as aircraft body pressure monitoring in aerospace and wind power generation.

[0061] To achieve the above objectives, this invention first designs the structure of a sensor data processing system, which mainly comprises three modules:

[0062] (1) Laser control module: In conjunction with the hardware features, it precisely controls the laser to emit laser light of a specified wavelength in a specified step size.

[0063] (2) FPGA acquisition module: The high-speed FPGA chip is used to control the AD sampling to obtain the original spectral data after photoelectric conversion of the sensor, and after certain data processing, it is sent to the processor.

[0064] (3) Algorithm module: After receiving the raw spectral data, different algorithms are selected according to different sensors to calculate the corresponding physical quantities.

[0065] The hardware mainly consists of a laser, a spectral standard, a photoelectric conversion board, an FPGA acquisition board, an algorithm CPU board, and an external communication board. It can connect to various fiber optic grating sensors and various MEMS optical sensors.

[0066] See Figure 1 This is a flowchart illustrating a multi-sensor data processing method provided in an embodiment of the present invention, as shown below. Figure 1 As shown in the figure, this embodiment of the invention illustrates the process of multi-sensor data processing by a processor.

[0067] Step S101: Group the monitoring data according to data elements to obtain multiple data groups, wherein the data elements include multiple physical parameters.

[0068] In this embodiment of the invention, the monitoring data can be the data acquired by the FPGA acquisition module. The data element can include three physical parameters: temperature, pressure, and wavelength. Of course, in specific implementations, in order to target more complex multi-sensor data processing systems, the data element can also include other physical parameters such as humidity and altitude. This embodiment of the invention is only an exemplary description.

[0069] Furthermore, the number of physical parameters is not limited to the three described in the embodiments of the present invention, but can be any other number. The multi-sensor data processing method of the present invention is applicable to curve, surface or higher-dimensional fitting.

[0070] To facilitate the explanation of the data processing method of the present invention, the embodiments of the present invention are described using surface fitting of three physical parameters: temperature, pressure, and wavelength as an example.

[0071] The monitoring data is described using the following formula:

[0072]

[0073] Among them, temperature ij =temperature ik ,preasure ij =preasure ik,j,k∈[1,n],j,k∈Z + Each row represents the wavelength under temperature and pressure conditions.

[0074] In an exemplary embodiment, the monitoring data can be represented as follows, wherein the data in the first row indicates that the wavelength obtained is 1555.367400nm when the temperature is 278.15K and the pressure is 70kpa, and the data in the other rows follow the same pattern, which will not be described in detail here.

[0075]

[0076] In the first implementation scenario, see Figure 2 Figure 1 is a flowchart illustrating a data grouping method provided in an embodiment of the present invention. As shown in the figure, the method includes:

[0077] Step S1011: Extract wavelengths from the monitoring data under the same temperature and pressure conditions to obtain multiple reference groups.

[0078] Wavelength data measured under the same temperature and pressure conditions are divided into multiple reference groups, which can be represented by the following formula:

[0079] A′=(A′1 A′2…A′ m )

[0080] in i = 1, 2, 3, ..., m. m is the number of groups obtained after grouping, and n is the number of times the wavelength is measured under the same temperature and pressure.

[0081] In the reference group, any reference group A′ i These represent multiple wavelengths collected under the same temperature and pressure conditions.

[0082] Step S1012: Extract data from the multiple reference groups in chronological order to obtain multiple data groups.

[0083] Furthermore, from the reference group obtained through grouping, data from the corresponding rows are extracted sequentially according to time order, and combined to obtain multiple data groups. The specific formula is as follows:

[0084] B = (B1 B2...B n )

[0085] in,

[0086] Specifically, due to each reference group A′ iThe data in the data can be understood as being collected chronologically. The first row of data in each reference group is extracted to form the corresponding B1, the second row of data in each reference group is extracted to form the corresponding B2, and so on to obtain all the data groups.

[0087] For a specific example, a dataset can be represented in the following form:

[0088]

[0089] Columns 1 to 3 represent B1, and the last 3 columns represent B. n .

[0090] In the second implementation scenario, see Figure 3 Figure 1 is a flowchart illustrating another data grouping method provided in an embodiment of the present invention. As shown in the figure, the method includes:

[0091] Step S1013: Extract wavelengths from the monitoring data under the same temperature and pressure conditions to obtain multiple reference groups.

[0092] The form of the reference group is as described in the above embodiments, and will not be repeated here.

[0093] Step S1014: Randomly extract data from the plurality of reference groups to obtain a plurality of data groups.

[0094] In this embodiment of the invention, from each reference group A′ i Randomly extract one row of data from each row to form a data group, thus ultimately obtaining n data groups B. i .

[0095] By randomly extracting data and randomly distributing it within the reference data set, the robustness of the data set can be effectively improved, thereby enhancing the accuracy and applicability of multi-sensor data computation.

[0096] In the third implementation scenario, the reference group data can be preprocessed to further eliminate interfering data, thereby improving data processing accuracy, reducing the amount of data processed, and increasing processing efficiency. See also Figure 4 Figure 1 is a flowchart illustrating a reference group data processing method provided in an embodiment of the present invention. As shown in the figure, the method includes:

[0097] Step S1015: Group the wavelengths under the same temperature and pressure conditions into multiple initial groups.

[0098] The monitoring data described in the above embodiments are also grouped according to the same temperature and pressure conditions to obtain multiple initial groups. The expression form of the initial groups is referred to as A′.

[0099] Step S1016: Calculate the wavelength dispersion in the initial group.

[0100] For each initial group, the dispersion of the corresponding wavelength is calculated. The dispersion can be calculated using methods such as range, mean deviation, variance, and standard deviation, which will not be elaborated further in this embodiment of the invention.

[0101] Step S1017: When the dispersion is greater than or equal to the first threshold, delete the data with the largest error in the initial group, and calculate the dispersion of the wavelength in the initial group after deletion.

[0102] A preset dispersion threshold is set. Taking the range calculation method as an example, the first threshold can be set to 1nm. Of course, in specific implementation, the setting range of the first threshold can be weighted according to the system accuracy and sampling environment, and is not limited to the setting range of the above embodiment.

[0103] When the calculated dispersion is greater than or equal to the first threshold, it indicates that the noise is relatively large. Further, the data with the largest error in the initial group is deleted, and the dispersion of the wavelength in the deleted initial group is recalculated.

[0104] Step S1018: When the dispersion is less than the first threshold, the initial group corresponding to the dispersion is used as the reference group.

[0105] When the calculated dispersion is less than the first threshold, it indicates that the noise is within the allowable range. The initial group at this time is then used as the reference group for subsequent processing to obtain the data group.

[0106] Step S102: Fit the multiple data groups separately, select the data with the smallest fitting error from each data group, and combine them to obtain the training set.

[0107] For each data group B i The fitting is performed separately. The fitting method can be the least squares method, but it is not limited to this method. Any method for surface fitting is within the protection scope of the embodiments of this invention.

[0108] After fitting, the calculation expressions for temperature, pressure, and wavelength are obtained. The fitting error between the fitted curve and each data point is then calculated using the following formula:

[0109] error = |y-y'|

[0110] Where y and y' represent the actual value and the predicted value, respectively.

[0111] Further select the data with the smallest fitting error, so that each data group B i We obtain one optimal data point from each sample, and combine them to obtain the training set, as shown in the following expression:

[0112]

[0113] To further improve the accuracy of the training set, please refer to the following for specific implementation: Figure 5 Figure 1 is a flowchart illustrating the first data filtering method provided in an embodiment of the present invention. As shown in the figure, the method includes:

[0114] Step S1021: Calculate the fitting error, which includes one or more of the following: mean square error, root mean square error, and mean absolute error.

[0115] Based on the description of the above embodiments, for any data group B i After the fitting is completed, its error can be calculated. This error can be one or more of the following: mean square error, root mean square error, and mean absolute error.

[0116] Step S1022: When the error is greater than or equal to the second threshold, delete the data with the largest error in the data group to obtain a new data group, and fit the new data group.

[0117] When the calculated error is greater than or equal to the preset second threshold, it indicates that the surface obtained by fitting at this time deviates significantly from the collected data, which is attributed to the error caused by noise. Further processing of the data set is performed, the data with the largest error is deleted, and multivariate surface fitting is performed on the new data set obtained after deletion.

[0118] Step S1023: When the error is less than the second threshold, select the data with the smallest fitting error from the data set.

[0119] When the calculated error is less than the preset second threshold, the characterization fitting surface matches the sampled data well, further eliminating noisy data. The resulting data set is then used in subsequent fitting calculations, and finally the optimal data set is selected to form the training set.

[0120] Step S103: Use the training set to fit and calculate the mathematical relationship between data elements.

[0121] Using the training set obtained from the above steps, the system of linear equations constructed according to the least squares method is shown in the following formula:

[0122]

[0123] in, It is the constant term of the system of linear equations.

[0124] The linear equations of temperature, pressure, and wavelength are solved using the Gaussian elimination method, and the solutions are stored in the sensor data processing system for subsequent calculations of pressure values.

[0125] Overall Fitting Performance Comparison: The multi-sensor data processing method of this invention can effectively improve the accuracy of data fitting. Traditional averaging methods and outlier removal averaging methods may increase data noise, thus failing to effectively improve fitting accuracy. Five pressure wavelength measurements were performed under the same pressure and temperature conditions. The comparison shows that the overall fitting accuracy of the averaging method and outlier removal averaging method is lower than that of the algorithm of this invention. (See [link to relevant documentation]). Figure 6 and Figure 7 .

[0126] Accuracy Comparison: Five pressure wavelengths were measured under the same pressure and temperature conditions. The comparison shows that the fitting accuracy of the averaging method and the outlier removal averaging method is lower than that of the algorithm provided in this invention. (See [link to relevant documentation]). Figure 8 .

[0127] To verify the reliability of the algorithm, under the same pressure and temperature conditions, this invention selected data four times from the original data, each time selecting a different amount of original data. The data were then averaged, and the fitting error was compared to demonstrate the reliability of the multi-sensor data processing method proposed in this invention. (See [link to relevant documentation]). Figure 9 .

[0128] As described in the above embodiments, the multivariate sensor data processing method provided by this invention first groups different wavelength data measured under the same temperature and pressure conditions into groups, and then performs multivariate surface fitting on the wavelength and temperature in each group of data. By calculating the fitting error of each group, the wavelength data that minimizes the fitting error is selected and then reassembled into a single data group. Finally, the reassembled data is used as a training set for fitting, and the final fitting coefficients are recorded for subsequent pressure value calculation. The resulting fitting is superior to traditional fitting methods. The multivariate wavelength-temperature-pressure fitting algorithm proposed in this invention has high adaptability and accuracy, and the robustness and accuracy of the fitting results can be improved through an automatic point elimination algorithm. This algorithm can be applied not only to wavelength-pressure correlation fitting scenarios but also to various sensor data fitting scenarios. Therefore, this algorithm has significant application value in sensor data processing and fitting, effectively improving fitting accuracy and robustness, reducing fitting errors, and bringing significant advantages and improvements to the field of sensor data analysis and application.

[0129] Through the description of the above method embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0130] This invention provides a non-volatile computer storage medium storing computer-executable instructions that can execute the multi-sensor data processing method in any of the above method embodiments.

[0131] Corresponding to the embodiment of the multi-sensor data processing method provided by the present invention, the present invention also provides a multi-sensor data processing device.

[0132] See Figure 10 Figure 1 is a schematic diagram of a multi-sensor data processing device provided in an embodiment of the present invention. As shown in the figure, the device includes:

[0133] Grouping module 11 is used to group monitoring data according to data elements to obtain multiple data groups, wherein the data elements include multiple physical parameters;

[0134] The training set generation module 12 is used to fit the multiple data groups respectively, select the data with the smallest fitting error from each data group, and combine them to obtain a training set.

[0135] The calculation module 13 is used to calculate the mathematical relationship between data elements by fitting the training set.

[0136] In an exemplary embodiment, the grouping module 11 can also be used to extract wavelengths under the same temperature and pressure conditions from the monitoring data when the data elements include wavelength, temperature and pressure, to obtain multiple reference groups; and to extract data from the multiple reference groups in chronological order or randomly to obtain multiple data groups.

[0137] In an exemplary embodiment, the grouping module 11 is further configured to: group wavelengths under the same temperature and pressure conditions into multiple initial groups; calculate the dispersion of wavelengths in the initial groups; when the dispersion is greater than or equal to a first threshold, delete the data with the largest error in the initial group and calculate the dispersion of wavelengths in the deleted initial group; when the dispersion is less than the first threshold, use the initial group corresponding to the dispersion as a reference group.

[0138] In an exemplary embodiment, the training set generation module 12 is further configured to: calculate the fitting error, the error including one or more of mean square error, root mean square error, and mean absolute error; when the error is greater than or equal to a second threshold, delete the data with the largest error in the data set to obtain a new data set, and fit the new data set; when the error is less than the second threshold, select the data with the smallest fitting error from the data set.

[0139] Figure 11 This is a schematic diagram of the structure of the multi-sensor data processing system provided in an embodiment of the present invention, as shown below. Figure 11 As shown, the system includes:

[0140] One or more processors 310 and memory 320, Figure 11 Take the 310 processor as an example.

[0141] The device for performing the multi-sensor data processing method may also include: a laser controller 330, a data acquisition unit 340, an input device 350, and an output device 360.

[0142] The laser controller 330, data acquisition unit 340, processor 310, memory 320, input device 350, and output device 360 ​​can be connected via a bus or other means. Figure 11 Taking the example of a connection between China and Israel via a bus.

[0143] Laser controller 330 is used to control the laser to emit laser light of a preset wavelength;

[0144] The data acquisition unit 340 is used to acquire the monitoring data of the sensor after photoelectric conversion and send it to the processor 310.

[0145] Memory 320, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the battery health state estimation method in this embodiment of the invention (e.g., attached...). Figure 10The diagram shows the grouping module 11, training set generation module 12, and calculation module 13. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby implementing the multi-sensor data processing method described in the above embodiment.

[0146] The memory 320 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the use of the processing device based on battery health estimation. Furthermore, the memory 320 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 320 may optionally include memory remotely located relative to the processor 310, and this remote memory may be connected to the battery health estimation processing device via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0147] The input device 350 can receive input digital or character information, and generate key signal inputs related to user settings and function control of the multi-sensor data processing device. The output device 360 ​​may include a display device such as a screen.

[0148] The one or more modules are stored in the memory 320, and when executed by the one or more processors 310, they perform the multi-sensor data processing method in any of the above method embodiments.

[0149] The above system can execute the methods provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the methods. Technical details not described in detail in this embodiment can be found in the methods provided in the embodiments of the present invention.

[0150] The electronic devices of this invention exist in various forms, including but not limited to:

[0151] (1) Mobile communication devices: These devices are characterized by their mobile communication capabilities and primarily aim to provide voice and data communication. These terminals include: smartphones (e.g., iPhones), multimedia phones, feature phones, and low-end phones, etc.

[0152] (2) Ultra-mobile personal computer devices: These devices fall under the category of personal computers, possessing computing and processing capabilities, and generally also have mobile internet access features. These terminals include PDAs, MIDs, and UMPCs, such as the iPad.

[0153] (3) Portable entertainment devices: These devices can display and play multimedia content. This category includes: audio and video players (such as iPods), handheld game consoles, e-books, as well as smart toys and portable car navigation devices.

[0154] (4) Server: A device that provides computing services. The components of a server include a processor, hard disk, memory, system bus, etc. Servers are similar to general computer architectures, but because they need to provide highly reliable services, they have higher requirements in terms of processing power, stability, reliability, security, scalability, and manageability.

[0155] (5) Other electronic devices with data interaction functions.

[0156] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0157] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for apparatus or system embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The apparatus and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0158] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0159] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for processing multi-sensor data, characterized in that, include: The monitoring data is grouped according to data elements to obtain multiple data groups, where the data elements include wavelength, temperature, and pressure. The multiple data sets are fitted separately, and the data with the smallest fitting error is selected from each data set and combined to obtain the training set. The mathematical relationships between data elements are obtained by fitting the training set.

2. The multi-sensor data processing method according to claim 1, characterized in that, The monitoring data is grouped according to data elements to obtain multiple data groups, including: Wavelengths under the same temperature and pressure conditions are extracted from the monitoring data to obtain multiple reference groups; Data is extracted from the multiple reference groups in chronological order to obtain multiple data groups.

3. The multi-sensor data processing method according to claim 1, characterized in that... The monitoring data is grouped according to data elements to obtain multiple data groups, including: Wavelengths under the same temperature and pressure conditions are extracted from the monitoring data to obtain multiple reference groups; Data is randomly extracted from the multiple reference groups to obtain multiple data groups.

4. The multi-sensor data processing method according to claim 2 or 3, characterized in that, The wavelengths obtained from the monitoring data under the same temperature and pressure conditions are used to obtain multiple reference groups, including: The wavelengths under the same temperature and pressure conditions are grouped into multiple initial groups; Calculate the wavelength dispersion in the initial group; When the dispersion is greater than or equal to the first threshold, delete the data with the largest error in the initial group, and calculate the wavelength dispersion in the initial group after deletion. When the dispersion is less than the first threshold, the initial group corresponding to the dispersion is used as the reference group.

5. The multi-sensor data processing method according to any one of claims 1 to 3, characterized in that, The step of selecting the data with the smallest fitting error from each data set includes: Calculate the fitting error, which includes one or more of the following: mean square error, root mean square error, and mean absolute error; When the error is greater than or equal to the second threshold, the data with the largest error in the data group is deleted to obtain a new data group, and the new data group is fitted. When the error is less than the second threshold, the data with the smallest fitting error is selected from the data set.

6. A multi-sensor data processing device, characterized in that, include: The grouping module is used to group the monitoring data according to data elements to obtain multiple data groups, wherein the data elements include wavelength, temperature and pressure; The training set generation module is used to fit the multiple data groups separately, select the data with the smallest fitting error from each data group, and combine them to obtain the training set. The calculation module is used to fit and calculate the mathematical relationships between data elements using the training set.

7. The multi-sensor data processing device according to claim 6, characterized in that, The grouping module is also used for, Wavelengths under the same temperature and pressure conditions are extracted from the monitoring data to obtain multiple reference groups; Data is extracted from the multiple reference groups in chronological order or randomly to obtain multiple data groups.

8. The multi-sensor data processing device according to claim 7, characterized in that, The grouping module is also used for, The wavelengths under the same temperature and pressure conditions are grouped into multiple initial groups; Calculate the wavelength dispersion in the initial group; When the dispersion is greater than or equal to the first threshold, delete the data with the largest error in the initial group, and calculate the wavelength dispersion in the initial group after deletion. When the dispersion is less than the first threshold, the initial group corresponding to the dispersion is used as the reference group.

9. The multi-sensor data processing apparatus according to any one of claims 6 to 8, characterized in that, The training set generation module is also used for, Calculate the fitting error, which includes one or more of the following: mean square error, root mean square error, and mean absolute error. When the error is greater than or equal to the second threshold, the data with the largest error in the data group is deleted to obtain a new data group, and the new data group is fitted. When the error is less than the second threshold, the data with the smallest fitting error is selected from the data set.

10. A multi-sensor data processing system, characterized in that, include: A laser controller is used to control a laser to emit laser light at a preset wavelength. The data acquisition unit is used to obtain the monitoring data from the sensor after photoelectric conversion and send it to the processor; A memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the multi-sensor data processing method as described in any one of claims 1 to 5.

Images