Calibration methods, devices, equipment and media for environmental parameter acquisition devices

Abstract

This application provides a calibration method, apparatus, equipment, and medium for an environmental parameter acquisition device, relating to the field of instrument calibration technology. In this application, firstly, multiple raw environmental parameter data corresponding to various types are acquired at a target time, and a target measured value vector is constructed based on these raw environmental parameter data; secondly, based on a first standard value vector, the target measured value vector undergoes a first calibration process to obtain a first calibration vector; then, based on the coupling correlation between various types, the first calibration vector undergoes a second calibration process to obtain a second calibration vector; further, based on a fixed mapping calibration matrix, the second calibration vector undergoes a third calibration process to obtain a third calibration vector; finally, based on the third calibration vector, multiple target calibration environmental parameter data are obtained. Based on the above, the relatively low reliability of parameter calibration in existing technologies can be improved.

Description

Technical Field

[0001] This application relates to the field of instrument calibration technology, and more specifically, to a calibration method, apparatus, equipment, and medium for an environmental parameter acquisition device. Background Technology

[0002] Currently, the calibration methods for multi-parameter environmental data loggers in the industry are relatively simple and fixed. The mainstream implementation schemes are mainly divided into two categories, and the specific implementation methods are as follows: The first type is conventional single-parameter fixed-model offline calibration, which is also the common technical solution for most equipment manufacturers in the industry. This solution only involves calibration during the equipment's factory shipment or periodic shutdown maintenance. Personnel use standard calibration equipment to perform single-point calibration of parameters such as temperature, humidity, and pressure. The second type is simple single-parameter online fine-tuning technology. Some high-end data acquisition devices are equipped with basic online calibration functions, but the overall implementation logic is relatively simple. It relies solely on historical data for a single parameter and uses the basic least squares method to make minor adjustments to the calibration coefficients. This solution can only simply compensate for the slow single-point drift of the sensor and is only suitable for scenarios where a single parameter is slowly decaying.

[0003] In other words, in the existing technology, there is a problem that the reliability of parameter calibration of environmental parameter acquisition devices is relatively low. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a calibration method, apparatus, device and medium for an environmental parameter acquisition device, so as to improve the problem of relatively low reliability of parameter calibration of environmental parameter acquisition devices in the prior art.

[0005] To achieve the above objectives, this application adopts the following technical solution: A calibration method for an environmental parameter acquisition device, comprising: Acquire multiple raw environmental parameter data of various types collected by the environmental parameter collector at the target time, and construct a target measured value vector based on the multiple raw environmental parameter data; Based on a pre-configured first standard value vector, the target measured value vector is subjected to a first calibration process to obtain a first calibration vector. The first standard value vector includes multiple standard environmental parameter data corresponding to the multiple types collected in historical time, and the first calibration vector includes multiple first calibration environmental parameter data corresponding to the multiple types. Based on the coupling relationship between the various types, the first calibration vector is subjected to a second calibration process to obtain a second calibration vector, wherein the second calibration vector includes multiple second calibration environment parameter data corresponding to the various types; Based on a predetermined solidified mapping calibration matrix, the second calibration vector is subjected to a third calibration process to obtain a third calibration vector. The solidified mapping calibration matrix is ​​a parameter that is iterated and solidified based on a pre-configured second standard value vector through a recursive least squares algorithm. The second standard value vector includes multiple standard environmental parameter data corresponding to the various types collected in historical time. The third calibration vector includes multiple third calibration environmental parameter data corresponding to the various types. Based on the third calibration vector, multiple target calibration environment parameter data corresponding to the multiple original environment parameter data are obtained, wherein there is a one-to-one correspondence between the multiple target calibration environment parameter data and the multiple original environment parameter data.

[0006] In a preferred embodiment of this application, in the calibration method for the aforementioned environmental parameter acquisition device, the step of performing a first calibration process on the target measured value vector based on a pre-configured first standard value vector to obtain a first calibration vector includes: Based on a pre-configured first standard value vector and historical measured value vector, each type of calibration model is fitted and determined to obtain a target calibration model corresponding to each type. The historical measured value vector includes multiple historical environmental parameter data corresponding to the various types, which are collected by the environmental parameter collector at the historical time corresponding to the first standard value vector. The target calibration model includes an offset calibration model, a proportional calibration model, a linear calibration model, or a quadratic polynomial calibration model, and the fitting and determination are based on the coefficient of determination and / or the mean absolute error. Based on the target calibration model corresponding to each type, the original environmental parameter data of the corresponding type in the target measured value vector are subjected to the first calibration process to obtain the first calibration vector.

[0007] In a preferred embodiment of this application, the step of performing a second calibration process on the first calibration vector based on the coupling relationship between the various types to obtain a second calibration vector includes: Based on the multiple historical environmental parameter data and constant terms corresponding to the various types collected by the environmental parameter collector in historical time, a target input vector is constructed. The target input vector is solved by batch least squares to obtain the target weight matrix and the target bias vector, wherein the target weight matrix and the target bias vector are used to characterize the coupling relationship between the various types; Based on the target weight matrix and the target bias vector, the first calibration vector is subjected to a second calibration process to obtain a second calibration vector.

[0008] In a preferred embodiment of this application, in the calibration method for the aforementioned environmental parameter acquisition device, the step of performing a third calibration process on the second calibration vector based on a predetermined fixed mapping calibration matrix to obtain a third calibration vector includes: Based on a predetermined fixed mapping calibration matrix, the second calibration vector is subjected to a third calibration process to obtain a third calibration result, wherein the third calibration result includes multiple third calibration environment parameter data corresponding to the various types; When the third calibration result is within the range of the environmental parameter acquisition device, the third calibration result is determined as the third calibration vector; When the result of the third calibration process is not within the range of the environmental parameter acquisition device, a new fixed mapping calibration matrix is ​​obtained by iterating and solidifying the result using a recursive least squares algorithm.

[0009] In a preferred embodiment of this application, the calibration method for the environmental parameter acquisition device further includes a step of forming the solidified mapping calibration matrix, which includes: Based on multiple second standard value vectors corresponding to multiple historical times, an initial matrix is ​​constructed, forming the initial coefficient vector and the initial covariance inverse matrix of the recursive least squares algorithm; Based on the environmental parameter data collected by the environmental parameter collector corresponding to the last historical time among the multiple historical times, and the output data after the first and second calibration processes, the current iteration input vector is constructed. Based on the second standard value vector corresponding to the last historical time among the multiple historical times, anomaly filtering judgment is performed on the current iteration input vector. When the current iteration input vector does not need to be filtered, the initial coefficient vector and the initial covariance inverse matrix are updated based on the current iteration input vector and the initial matrix to obtain the updated coefficient vector and the updated covariance inverse matrix. The updated coefficient vector and the updated covariance inverse matrix are used as the objects of the next update. Based on the updated coefficient vector, the solidification mapping calibration matrix is ​​determined.

[0010] In a preferred embodiment of this application, the step of determining the solidified mapping calibration matrix based on the updated coefficient vector in the above-mentioned environmental parameter acquisition device calibration method includes: Based on the updated coefficient vector, the current iteration input vector is mapped, and the mapping result is adjusted to the second standard value vector corresponding to the last historical time in the plurality of historical times according to a preset convergence strength, so as to obtain the current iteration output vector. The solidification is determined based on the current iteration output vector, and when it is determined that solidification is possible, the updated coefficient vector is determined as the solidification mapping calibration matrix.

[0011] In a preferred embodiment of this application, in the calibration method for the aforementioned environmental parameter acquisition device, the step of determining the solidification based on the current iterative output vector, and determining the updated coefficient vector as the solidification mapping calibration matrix when it is determined that solidification is possible, includes: At least the current iteration output vector is subjected to range constraint judgment and rate of change constraint judgment; When the current iterative output vector satisfies the range constraint judgment condition and the rate of change constraint judgment condition, the current iterative output vector is determined as a stable output vector, and the current stable duration is obtained. When the current iterative output vector does not satisfy the range constraint judgment condition or the rate of change constraint judgment condition, the current iterative output vector is not determined as a stable output vector, and the current stable duration is cleared. The stable duration is used to characterize the duration from the first determination of the stable output vector after the clearing process to the current time. Based on the current stable duration, determine whether the preset curing conditions are met, and if the preset curing conditions are met, determine the updated coefficient vector as the curing mapping calibration matrix.

[0012] This application also provides a calibration device for an environmental parameter acquisition device, comprising: The measured value vector construction module is used to acquire multiple raw environmental parameter data of various types collected by the environmental parameter collector at the target time, and to construct a target measured value vector based on the multiple raw environmental parameter data. The first calibration processing module is used to perform a first calibration process on the target measured value vector based on a pre-configured first standard value vector to obtain a first calibration vector. The first standard value vector includes multiple standard environmental parameter data corresponding to the multiple types collected in historical time, and the first calibration vector includes multiple first calibration environmental parameter data corresponding to the multiple types. The second calibration processing module is used to perform a second calibration processing on the first calibration vector based on the coupling relationship between the multiple types to obtain a second calibration vector, wherein the second calibration vector includes multiple second calibration environment parameter data corresponding to the multiple types; The third calibration processing module is used to perform a third calibration process on the second calibration vector based on a pre-determined solidified mapping calibration matrix to obtain a third calibration vector. The solidified mapping calibration matrix is ​​a parameter that is iterated and solidified based on a pre-configured second standard value vector through a recursive least squares algorithm. The second standard value vector includes multiple standard environmental parameter data corresponding to the various types collected in historical time. The third calibration vector includes multiple third calibration environmental parameter data corresponding to the various types. The calibration data determination module is used to obtain multiple target calibration environment parameter data corresponding to the multiple original environment parameter data based on the third calibration vector, wherein there is a one-to-one correspondence between the multiple target calibration environment parameter data and the multiple original environment parameter data.

[0013] Based on the above, this application also provides an electronic device, including: Memory, used to store computer programs; A processor connected to the memory is used to execute the computer program stored in the memory to implement the above-described calibration method for the environmental parameter acquisition device.

[0014] Based on the above, this application also provides a computer-readable storage medium storing a computer program that, when executed, performs the various steps of the calibration method for the environmental parameter acquisition device described above.

[0015] The calibration method, apparatus, equipment, and medium for the environmental parameter acquisition device provided in this application firstly acquire multiple raw environmental parameter data corresponding to various types collected at a target time, and construct a target measured value vector based on the multiple raw environmental parameter data; secondly, perform a first calibration process on the target measured value vector based on a first standard value vector to obtain a first calibration vector; then, perform a second calibration process on the first calibration vector based on the coupling correlation between various types to obtain a second calibration vector; further, perform a third calibration process on the second calibration vector based on a fixed mapping calibration matrix to obtain a third calibration vector; finally, obtain multiple target calibration environmental parameter data based on the third calibration vector. Based on the above, on the one hand, the first calibration process enables calibration based on standard environmental parameter data, which can reduce acquisition deviations caused by inherent sensor errors and hardware channel differences to a certain extent. On the other hand, the second calibration process enables calibration based on coupling correlation, which can offset interference errors caused by dynamic environmental changes to a certain extent, effectively improving the data consistency of multi-parameter synchronous acquisition. Furthermore, the third calibration process enables calibration based on a fixed mapping calibration matrix, which can improve the long-term accuracy drift caused by sensor aging and gradual changes in operating conditions to a certain extent, achieving dynamic calibration throughout the equipment's lifecycle. Therefore, by combining these three calibration processes, comprehensive data calibration can be achieved, obtaining reliable target calibration environmental parameter data, thereby improving the problem of low reliability in parameter calibration in existing technologies. Attached Figure Description

[0016] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings.

[0017] Figure 1 A structural block diagram of an electronic device provided in an embodiment of this application.

[0018] Figure 2 This is a flowchart illustrating the calibration method for an environmental parameter acquisition device provided in an embodiment of this application.

[0019] Figure 3 A block diagram of a calibration device for an environmental parameter acquisition device provided in an embodiment of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0021] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] like Figure 1 As shown in the illustration, this application provides an electronic device. The electronic device may include a memory, a processor, and a calibration device for an environmental parameter acquisition unit.

[0023] Specifically, the memory and the processor are electrically connected directly or indirectly to enable data transmission or interaction. For example, the memory and the processor can be electrically connected via one or more communication buses or signal lines. The calibration device for the environmental parameter acquisition device includes at least one software functional module stored in the memory in the form of software or firmware. The processor is used to execute executable computer programs stored in the memory, such as the software functional modules and computer programs included in the calibration device for the environmental parameter acquisition device, to implement the calibration method for the environmental parameter acquisition device provided in this application embodiment.

[0024] Optionally, the memory may be, but is not limited to, random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), etc.

[0025] Optionally, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a system on chip (SoC), etc.; it can 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 gate or transistor logic devices, or discrete hardware components.

[0026] Understandable, Figure 1 The structure shown is for illustrative purposes only; the electronic device may also include components that are more advanced than those shown. Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown may include, for example, a communication unit for exchanging information with other devices.

[0027] Combination Figure 2 This application also provides a calibration method for an environmental parameter acquisition device applicable to the aforementioned electronic device. The method steps defined in the calibration method for the environmental parameter acquisition device can be implemented by the electronic device.

[0028] The following will be about Figure 2 The specific process shown will be explained in detail.

[0029] Step S110: Obtain multiple raw environmental parameter data corresponding to various types collected by the environmental parameter collector at the target time, and construct a target measured value vector based on the multiple raw environmental parameter data.

[0030] In this embodiment, the electronic device can acquire multiple raw environmental parameter data of various types collected by the environmental parameter acquisition device at a target time, and construct a target measured value vector based on the multiple raw environmental parameter data. For example, the environmental parameter acquisition device operates normally upon power-on, and can collect eight raw environmental parameter data in real time: temperature, pressure, pressure difference, oxygen concentration, relative humidity, dew point, absolute humidity, and particle number. Noise reduction and amplification preprocessing are performed to ensure the raw signal is clean and effective. Then, a target measured value vector can be constructed, such as... m0 represents the first raw environmental parameter data, such as temperature; m1 represents the second raw environmental parameter data, such as pressure; and m7 represents the eighth raw environmental parameter data, such as particle number.

[0031] Step S120: Based on the pre-configured first standard value vector, perform a first calibration process on the target measured value vector to obtain a first calibration vector.

[0032] In this embodiment, after obtaining the target measured value vector, the electronic device can perform a first calibration process on the target measured value vector based on a pre-configured first standard value vector to obtain a first calibration vector. The first standard value vector includes multiple standard environmental parameter data corresponding to the various types collected over historical time. The first calibration vector includes multiple first calibration environmental parameter data corresponding to the various types (e.g., from a standard, calibration source, or known true value). The first standard value vector can be represented as... S0 represents the first standard environmental parameter data, such as temperature; S1 represents the second standard environmental parameter data, such as pressure; and S7 represents the eighth standard environmental parameter data, such as particle number.

[0033] Step S130: Based on the coupling relationship between the various types, perform a second calibration process on the first calibration vector to obtain a second calibration vector.

[0034] In this embodiment, after obtaining the second calibration vector, the electronic device can perform a second calibration process on the first calibration vector based on the coupling relationship between the various types (e.g., the influence of temperature on dew point, the influence of temperature and pressure combination on absolute humidity, the influence of relative humidity on oxygen concentration, the influence of pressure on particle number, etc.) to obtain the second calibration vector. The second calibration vector includes multiple second calibration environmental parameter data corresponding to the various types.

[0035] Step S140: Based on the predetermined solidified mapping calibration matrix, the second calibration vector is subjected to a third calibration process to obtain a third calibration vector.

[0036] In this embodiment, after obtaining the second calibration vector, the electronic device can perform a third calibration process on the second calibration vector based on a pre-determined fixed mapping calibration matrix to obtain a third calibration vector. The fixed mapping calibration matrix consists of parameters that are iterated and fixed based on a pre-configured second standard value vector using a recursive least squares (RLS) algorithm. The second standard value vector includes multiple standard environmental parameter data corresponding to the various types collected over historical time. The third calibration vector includes multiple third calibration environmental parameter data corresponding to the various types. It should be noted that the second standard value vector can be the same as or different from the first standard value vector, and can be selected according to actual needs. In this embodiment, they can be the same. When multiple second standard value vectors and multiple first standard value vectors are included, the multiple second standard value vectors and multiple first standard value vectors can be completely identical or partially identical.

[0037] Step S150: Based on the third calibration vector, obtain multiple target calibration environment parameter data corresponding to the multiple original environment parameter data.

[0038] In this embodiment, after obtaining the third calibration vector, the electronic device can obtain multiple target calibration environment parameter data corresponding to the multiple original environmental parameter data based on the third calibration vector. There is a one-to-one correspondence between the multiple target calibration environment parameter data and the multiple original environmental parameter data. For example, the multiple third calibration environment parameter data in the third calibration vector can be directly used as the multiple target calibration environment parameter data; alternatively, the multiple third calibration environment parameter data can be further processed, such as by using other calibration methods, to obtain the multiple target calibration environment parameter data.

[0039] Based on the above, on the one hand, the first calibration process enables calibration based on standard environmental parameter data, which can reduce acquisition deviations caused by inherent sensor errors and hardware channel differences to a certain extent. On the other hand, the second calibration process enables calibration based on coupling correlation, which can offset interference errors caused by dynamic environmental changes to a certain extent, effectively improving the data consistency of multi-parameter synchronous acquisition. Furthermore, the third calibration process enables calibration based on a fixed mapping calibration matrix, which can improve the long-term accuracy drift caused by sensor aging and gradual changes in operating conditions to a certain extent, achieving dynamic calibration throughout the equipment's lifecycle. Therefore, by combining these three calibration processes, comprehensive data calibration can be achieved, obtaining reliable target calibration environmental parameter data, thereby improving the problem of low reliability in parameter calibration in existing technologies.

[0040] It should be noted that the specific method of the first calibration process is not limited and can be selected according to actual needs.

[0041] For example, in an alternative implementation, in order to achieve a reliable first calibration process and effectively improve the problems of sensor inherent error, range nonlinearity error, zero drift, temperature drift and poor consistency of multiple acquisition channels, the above step S120 may further include the following steps S121 and S122.

[0042] Step S121: Based on the pre-configured first standard value vector and historical measured value vector, fit and determine the calibration model for each type to obtain the target calibration model corresponding to each type.

[0043] In this embodiment, a target calibration model can be obtained by fitting and determining each type of calibration model based on a pre-configured first standard value vector and historical measured value vector. The historical measured value vector includes multiple historical environmental parameter data corresponding to the various types, collected by the environmental parameter collector at the historical time corresponding to the first standard value vector. The target calibration model includes an offset calibration model, a proportional calibration model, a linear calibration model, or a quadratic polynomial calibration model. The fitting and determination criteria include the coefficient of determination and / or mean absolute error. For example, in ascending order of complexity (i.e., offset calibration model, proportional calibration model, linear calibration model, or quadratic polynomial calibration model), four types of calibration models are tried sequentially. The first calibration model that simultaneously satisfies the conditions related to the coefficient of determination and mean absolute error is selected as the target calibration model. If both conditions cannot be met simultaneously, the calibration model with the smallest mean absolute error is selected as the target calibration model. For example, a corresponding calibration point set can be constructed for each type based on the first standard value vector and historical measured value vector over N time periods, such as... ; , where p represents the type.

[0044] Step S122: Based on the target calibration model corresponding to each type, perform a first calibration process on the original environmental parameter data of the corresponding type in the target measured value vector to obtain a first calibration vector.

[0045] In this embodiment of the application, after obtaining the target calibration model, the original environmental parameter data of the corresponding type in the target measured value vector can be subjected to a first calibration process based on the target calibration model corresponding to each type to obtain a first calibration vector.

[0046] It should be noted that the specific method of the second calibration process is not limited and can be selected according to actual needs.

[0047] For example, in an alternative implementation, in order to achieve a reliable second calibration process that can offset interference errors caused by dynamic environmental changes and effectively improve the data consistency of multi-parameter synchronous acquisition, the above step S130 may further include the following steps S131, S132 and S133.

[0048] Step S131: Based on the multiple historical environmental parameter data and constant terms corresponding to the various types collected by the environmental parameter collector in historical time, construct the target input vector.

[0049] In this embodiment, a target input vector can be constructed based on multiple historical environmental parameter data and constant terms corresponding to the various types collected by the environmental parameter collector over historical time. It should be noted that the parameters in the target input vector are the results of the historical environmental parameter data after a first calibration process, such as constructing a 9-dimensional input vector (8 terms of the first calibration process output + constant term 1): .

[0050] Step S132: Solve the target input vector using the batch least squares method to obtain the target weight matrix and the target bias vector.

[0051] In this embodiment, after obtaining the target input vector, the target input vector can be solved using batch least squares to obtain the target weight matrix and the target bias vector. The target weight matrix and the target bias vector are used to characterize the coupling relationship between the various types. For example, an 8*8 weight matrix W and an 8-dimensional bias vector B can be obtained.

[0052] Step S133: Based on the target weight matrix and the target bias vector, perform a second calibration process on the first calibration vector to obtain a second calibration vector.

[0053] In this embodiment of the application, after obtaining the target weight matrix and the target bias vector, a second calibration process can be performed on the first calibration vector based on the target weight matrix and the target bias vector to obtain a second calibration vector. For example, ,in, For the first calibration vector, This is the second calibration vector.

[0054] It should be noted that the specific method of the third calibration process is not limited and can be selected according to actual needs.

[0055] For example, in an alternative implementation, in order to achieve a reliable third calibration process and improve the long-term accuracy drift caused by sensor aging and gradual changes in operating conditions, the above step S140 may further include steps S141, S142 and S143.

[0056] Step S141: Based on the predetermined solidified mapping calibration matrix, perform a third calibration process on the second calibration vector to obtain the third calibration result.

[0057] In this embodiment, the second calibration vector can be subjected to a third calibration process based on a pre-determined fixed mapping calibration matrix to obtain a third calibration result. The third calibration result includes multiple third calibration environment parameter data corresponding to the various types. For example, the fixed mapping calibration matrix and the second calibration vector can be multiplied to obtain the third calibration result. Additionally, the fixed mapping calibration matrix may also include weights and biases.

[0058] Step S142: When the third calibration result is within the range of the environmental parameter acquisition device, the third calibration result is determined as the third calibration vector.

[0059] In this embodiment of the application, after obtaining the third calibration result, the third calibration result can be determined as the third calibration vector when it is within the range of the environmental parameter acquisition device.

[0060] Step S143: When the third calibration result is not within the range of the environmental parameter acquisition device, a new fixed mapping calibration matrix is ​​obtained by iterating and solidifying the result using a recursive least squares algorithm.

[0061] After obtaining the third calibration result, if the third calibration result is not within the range of the environmental parameter acquisition device, a new fixed mapping calibration matrix is ​​obtained by iterating and solidifying through a recursive least squares algorithm (the iteration and solidification process is described later).

[0062] Based on the above embodiments, in order to ensure that the solidified mapping calibration matrix has high reliability, the calibration method of the environmental parameter acquisition device may further include the step of forming the solidified mapping calibration matrix. This step may include steps S160, S170, S180 and S190, and the specific contents of each step are as follows.

[0063] Step S160: Based on multiple second standard value vectors corresponding to multiple historical times, construct an initial matrix and form the initial coefficient vector and initial covariance inverse matrix of the recursive least squares algorithm.

[0064] In this embodiment, an initial matrix can be constructed based on multiple second standard value vectors corresponding to multiple historical times (for example, an 8*8 initial matrix can be constructed based on 8 second standard value vectors corresponding to 8 historical times), forming the initial coefficient vector and the initial covariance inverse matrix of the recursive least squares algorithm, that is, initializing the RLS state for each output parameter: initializing the weight coefficient vector. Initialize the inverse covariance matrix. .

[0065] Step S170: Based on the output data of the multiple historical environmental parameter data of the multiple types corresponding to the environmental parameter collectors corresponding to the last historical time in the multiple historical times, after the first calibration processing and the second calibration processing, the current iteration input vector is constructed.

[0066] In this embodiment of the application, the current iteration input vector can also be constructed based on the output data of the multiple historical environmental parameter data of the multiple types corresponding to the environmental parameter collectors collected from the last historical time in the multiple historical time, after the first calibration processing and the second calibration processing. For example, 8-dimensional output data + constant 1.

[0067] Step S180: Based on the second standard value vector corresponding to the last historical time among the multiple historical times, perform anomaly filtering judgment on the current iteration input vector; and when the current iteration input vector does not need to be filtered, update the initial coefficient vector and the initial covariance inverse matrix based on the current iteration input vector and the initial matrix to obtain the updated coefficient vector and the updated covariance inverse matrix.

[0068] In this embodiment, after obtaining the current iteration input vector, anomaly filtering can be performed on the current iteration input vector based on the second standard value vector corresponding to the last historical time among the multiple historical times. Furthermore, when the current iteration input vector does not need to be filtered (e.g., the difference between the parameter in the current iteration input vector and the corresponding parameter in the second standard value vector is less than an anomaly threshold, such as the full scale of the parameter), the initialization coefficient vector and the initialization covariance inverse matrix are updated based on the current iteration input vector and the initial matrix to obtain the updated coefficient vector and the updated covariance inverse matrix. The updated coefficient vector and the updated covariance inverse matrix are used as the objects for the next update. It should be noted that the iterative update process can refer to the relevant explanations of the recursive least squares algorithm, and will not be specifically limited or described here.

[0069] Step S190: Determine the solidified mapping calibration matrix based on the updated coefficient vector.

[0070] In this embodiment of the application, after obtaining the updated coefficient vector, the solidified mapping calibration matrix can be determined based on the updated coefficient vector.

[0071] It should be further explained that the specific method for determining the solidified mapping calibration matrix is ​​not limited. For example, in an alternative embodiment, in order to ensure the reliability of the solidified mapping calibration matrix, the above-mentioned step S190 may further include steps S191 and S192, wherein the specific contents of each step are as follows.

[0072] Step S191: Based on the updated coefficient vector, map the current iteration input vector, and, according to a preset convergence strength, converge and correct the mapping result to the second standard value vector corresponding to the last historical time among the multiple historical times, to obtain the current iteration output vector.

[0073] In this embodiment, the current iteration input vector can be mapped based on the updated coefficient vector, and the mapping result can be adjusted to approximate the second standard value vector corresponding to the last historical time among the multiple historical times according to a preset approximation strength, to obtain the current iteration output vector. For example, the updated coefficient vector can be multiplied by the current iteration input vector (or multiplied by the weight and then added to the bias) to obtain the mapping result, and then approximation adjustment can be performed. ,in, For the i-th parameter in the current iteration output vector, For the i-th parameter in the mapping result, The preset approach strength is (e.g., 0.6). is the i-th parameter in the second standard value vector.

[0074] Step S192: Based on the current iteration output vector, solidification determination is performed, and when it is determined that solidification is possible, the updated coefficient vector is determined as the solidification mapping calibration matrix.

[0075] In this embodiment of the application, after obtaining the current iteration output vector, a solidification determination can be performed based on the current iteration output vector, and when it is determined that solidification is possible, the updated coefficient vector is determined as the solidification mapping calibration matrix.

[0076] It is understood that the specific method of solidification determination in step S192 above is not limited. For example, in an alternative implementation, in order to ensure that the obtained solidification mapping calibration matrix has high reliability, step S192 above may further include steps S192a, S192b and S192c, wherein the specific contents of each step are as follows.

[0077] Step S192a: At least the range constraint judgment and the rate of change constraint judgment are performed on the current iteration output vector.

[0078] In this embodiment, at least the current iterative output vector is subject to range constraint judgment and rate of change constraint judgment. Range constraint judgment refers to determining whether the parameters in the current iterative output vector belong to a corresponding range, i.e., greater than or equal to the lower limit and less than or equal to the upper limit. Rate of change constraint judgment refers to whether the difference between the current iterative output vector and the iterative output vector of the previous stage is less than or equal to the difference between the upper and lower limits. Additionally, in some embodiments, range constraint judgment and rate of change constraint judgment can also be performed on the parameters in the second standard value vector.

[0079] Step S192b: When the current iterative output vector satisfies the range constraint judgment condition and the rate of change constraint judgment condition, the current iterative output vector is determined as a stable output vector, and the current stable duration is obtained. When the current iterative output vector does not satisfy the range constraint judgment condition or the rate of change constraint judgment condition, the current iterative output vector is not determined as a stable output vector, and the current stable duration is cleared.

[0080] In this embodiment, after determining the range constraint and rate of change constraint, if the current iterative output vector satisfies both conditions, it is determined as a stable output vector, and the current stable duration is obtained. If the current iterative output vector does not satisfy either the range constraint or rate of change constraint, it is not determined as a stable output vector, and the current stable duration is reset to zero. The stable duration characterizes the time elapsed since the first stable output vector was determined after the reset process.

[0081] Step S192c: Based on the current stable duration, determine whether the preset curing conditions are met, and if the preset curing conditions are met, determine the updated coefficient vector as the curing mapping calibration matrix.

[0082] In this embodiment of the application, after obtaining the current stable duration, it can be determined whether a preset curing condition is met based on the current stable duration. If the preset curing condition is met, the updated coefficient vector is determined as the curing mapping calibration matrix. For example, if the current stable duration is greater than or equal to a preset time length, it is determined that the preset curing condition is met; if the current stable duration is less than a preset time length, it is determined that the preset curing condition is not met.

[0083] To facilitate understanding of the calibration method for the environmental parameter acquisition device described above, this application embodiment also provides a specific application example, the details of which are as follows.

[0084] 1.1.1 Data Access and Snapshot Processing The automatic calibration system receives a complete set of environmental parameter data, called a calibration snapshot, which includes: Standard value vector : From a standard, calibration source, or known truth value; Measured value vector : Raw readings from each channel of the data acquisition unit.

[0085] It accumulates multiple sets of historical snapshots and supports real-time push of current frame data in dynamic mode. It only participates in calibration calculations when all eight parameters are valid.

[0086] 1.1.2 First Layer: Basic Calibration Layer (L1) This layer forms the foundation of the entire calibration system, primarily addressing issues such as inherent sensor errors, range nonlinearity errors, zero-point drift, temperature drift, and poor consistency across multiple acquisition channels. This application abandons the traditional fixed-algorithm calibration mode, incorporating four universal calibration models—offset, proportional, linear, and polynomial—that can autonomously adapt to the optimal solution based on parameter characteristics. During the equipment calibration phase, the system acquires raw data from multiple points across the entire measurement range and their corresponding standard true values, performing fitting calculations for each of the four models. By comparing the fitting residuals, it automatically selects the calibration method with the best fit for the current parameters and range. The actual adaptation logic aligns with the inherent characteristics of the hardware: for parameters with good linearity such as pressure and differential pressure, a linear calibration model is prioritized; for parameters with significant nonlinear characteristics such as temperature, humidity, and dew point, a second-order or higher polynomial model is automatically matched; offset calibration is used for conditions with large sensor zero drift; and proportional calibration is employed for conditions with hardware gain deviations. Simultaneously, using the output signal of a high-precision standard device as a benchmark, the gain and offset parameters of all acquisition channels are uniformly calibrated, eliminating channel deviations caused by hardware batch differences and ensuring that all channel output data is consistent, standardized, and accurate.

[0087] 1.1.2.1 Implementation Steps Step S101: Adjust parameters Extract calibration point sets from historical snapshots: ; Step S102: Try the four types of calibration models in ascending order of complexity:

[0088] Step S103: For each candidate model, calculate the goodness-of-fit index: Coefficient of determination (Default requirement) ); Mean Absolute Error (Default requirement) ,in, (Standard value span).

[0089] Step S104: Select the first one that simultaneously satisfies The model is compared with the error threshold; if none of them are satisfied, the model with the smallest average error is selected as the downgrade.

[0090] Step S105: Calculate the original measured values ​​for the current frame. Apply the selected model parameter by parameter to obtain the L1 output vector. .

[0091] 1.1.2.2 Technical Features All parameters have a unified range and a single model, without segmentation by sub-intervals; All eight parameters use the same automatic mode selection process to achieve standardized calibration of multiple channels; The L1 output serves as the input feature vector for L2.

[0092] 1.1.3 Second Layer: Cross-compensation Layer (L2) This layer addresses the common multi-parameter coupling interference problem in the industry and is one of the core innovations that distinguishes this application from traditional single-parameter calibration schemes. Extensive field testing has verified that temperature, pressure, relative humidity, oxygen concentration, and particle number are the five core variables affecting environmental monitoring accuracy. Parameters such as dew point and absolute humidity are derived calculation parameters and are highly susceptible to fluctuations in these five parameters. Based on this, this application selects five core parameters to construct real-time feature vectors. Through numerous standard operating condition calibration experiments, the coupling correlation coefficients between each parameter are statistically analyzed and trained to construct a multi-dimensional cross-compensation matrix. This matrix can accurately quantify the mutual interference weights of various parameters, including the influence of temperature on dew point, the influence of temperature and pressure combinations on absolute humidity, the influence of humidity on oxygen concentration, and the influence of pressure on particle number. During equipment operation, the five-parameter data after basic calibration are retrieved in real time, and the coupling compensation amount of each parameter is accurately calculated through matrix operations to offset interference errors caused by dynamic environmental changes, effectively improving the data consistency of synchronously acquired multi-parameter data.

[0093] 1.1.3.1 Implementation Steps Step S201: Transform the measured values ​​of the historical snapshots using L1 to obtain the L1 snapshot set.

[0094] Step S202: Construct a 9-dimensional input vector (8 L1 output terms + constant term 1):

[0095] Step S203: For each output parameter Solve using batch least squares: This yields an 8×8 weight matrix W (composed of each...) Composition) and 8-dimensional bias vector B (composed of) constitute), The desired response belongs to the information that will ultimately be fitted, while and These are parameters that need to be learned.

[0096] Step S204: When the number of snapshots Enable full cross model when; When it is downgraded to a diagonally independent linear model; when L2 is not enabled at this time.

[0097] Step S205: The L2 matrix adopts a pre-trained fixed strategy—it is saved after batch fitting is completed, and refitting is only performed when the number of new snapshots increases. It is not overwritten by L3 during the running process.

[0098] Step S206: During inference, the L2 output is calculated as follows: .

[0099] 1.1.3.2 Technical Features off-diagonal elements Quantization parameters For parameters Cross-interference weights; L2 and L3 have separate responsibilities: L2 is responsible for batch cross-compensation, and L3 is responsible for online timing fine-tuning.

[0100] 1.1.4 Third Layer: Online Adaptive Layer (L3) This level is the top-level optimization unit, primarily addressing the long-term accuracy drift caused by sensor aging and gradual changes in operating conditions, achieving dynamic calibration throughout the equipment's lifecycle. This application uses cross-compensated calibration data as observations and industry standard true values ​​as target values, employing the RLS recursive least squares algorithm to continuously iteratively update the basic model parameters and cross-matrix weights. Compared to traditional least squares methods, the RLS algorithm has higher iteration efficiency, consumes fewer hardware resources, and is suitable for embedded equipment operation scenarios, continuously optimizing calibration parameters as operating conditions slowly change. Simultaneously, to avoid weight misalignment caused by abnormal interference such as instantaneous airflow fluctuations and short-term temperature and pressure changes, this application adds a stability judgment mechanism, setting reasonable data fluctuation thresholds and weight change thresholds. When instantaneous abnormal fluctuations occur in the data, the system pauses weight updates, retaining historically optimal calibration parameters; when the data is stable over a long period and the weights tend to converge, the calibration parameters are automatically fixed and updated, ensuring a smooth and reliable calibration process and allowing the equipment's long-term operating data to continuously conform to the standard true values.

[0101] 1.1.4.1 Implementation Steps initialization Step S301: After L2 is ready, the initial 8×8 matrix is ​​obtained by batch fitting the L2 output snapshot, which is used as the initial value of L3 coefficient (warm start).

[0102] Step S302: For each output parameter Initialize RLS state: Weight coefficient vector ; Inverse covariance matrix ; Online RLS iteration (per frame).

[0103] Step S303: Construct L2 current frame input (L2 output is 8-dimensional + constant 1).

[0104] Step S304: Anomaly Filtering – Calculate for each parameter: ; like ( For the first If the abnormal threshold (which defaults to the full scale of this parameter) is set, then the RLS update for this frame will be rejected.

[0105] Step S305: For frames that are not rejected, for each output parameter Execution with forgetting factor RLS updates (default 0.98): Kalman Gain: ; Prior error: ; Weight update: ; Covariance inverse matrix update: .

[0106] in, This indicates an update. Other parameters can be found in the relevant explanations of the RLS algorithm. This application directly utilizes the RLS algorithm without making any improvements.

[0107] Step S306: Synchronously update the L3 model matrix for inference and evaluation.

[0108] Calibration phase output convergence Step S307: In the calibration stage where the material has not yet cured and a standard true value exists, the final output can be determined by the approximation strength. (Default 0.6) Correct to standard value: ; In both solidified and pure inference modes, only the L3 matrix transformation result is output, without convergence.

[0109] 1.1.5 Stability Control and Parameter Fixation 1.1.5.1 Stability Criteria The measured values ​​and standard values ​​of the eight parameters simultaneously satisfy: Condition C1 (range constraint): ; in, For the first The upper and lower limits of the range configured for this parameter.

[0110] Condition C2 (rate of change constraint) (relative to the previous sampling frame): ; ; If all 8 conditions C1 and C2 are met, the current frame is determined to be a stable frame.

[0111] 1.1.6 Continuous and stable timing and curing 1.1.6.1 Implementation Steps Step S401: Start / continue the stabilization timer when a stable frame is obtained.

[0112] Step S402: When an unstable frame occurs, reset the timer and start accumulating again from zero.

[0113] Step S403: When the stability condition is continuously met for 10 minutes (600,000 ms): Copy the current L3 weight matrix to the fixed model; Persistently save to non-volatile storage (such as JSON / Flash / EEPROM); Set a solidification flag to stop RLS iteration; Subsequent inference uses fixed parameters.

[0114] 1.1.6.1 Technical Features Stability determination is based on the physical stability of the measured value and the standard value, rather than solely on the weight change rate; data stream interruption (silence) does not interrupt stability timing, and accumulation continues as long as the current frame still meets the conditions; the status bar / log can output the progress information "Stability timing within range: N / 10 minutes".

[0115] 1.1.7 Unattended Loop and Drift Recovery 1.1.7.1 Implementation Steps Step S501: Start the unattended dynamic mode and execute the L1→L2→L3 calibration process in a loop according to the configured cycle (e.g., every 2 seconds).

[0116] Step S502: After solidification, continuously monitor whether the measured value of the current frame and the standard value are still within the range.

[0117] Step S503: If the measurement range is exceeded or an abnormal operating condition is detected: Remove the fixed flag; reset the stable timing; restart the L3 online RLS iteration and stable timing; achieve automatic recovery of online calibration without manual intervention.

[0118] The technical solution of this application will be further described in detail below with reference to actual engineering application scenarios. This embodiment is only an exemplary application case and is not intended to limit the scope of protection of this application. All equivalent improvements and adaptations based on this solution are within the scope of protection of this application.

[0119] 1.1 Implementation Scenarios This embodiment is applied to environmental monitoring in a biopharmaceutical cleanroom. It employs an integrated multi-parameter environmental data acquisition device to simultaneously collect eight environmental parameters: temperature, pressure, differential pressure, oxygen concentration, relative humidity, absolute humidity, dew point, and particle count. The equipment operates online 24 / 7, and the workshop conditions involve normalized small fluctuations in temperature, humidity, and pressure. The sensors will slowly age and drift over time, requiring extremely high long-term stability and accuracy of the monitoring data.

[0120] 1.2 Specific Implementation Steps Step 1: Initialize Basic Adaptive Calibration. Upon initial deployment, the basic calibration process is automatically initiated. Utilizing standard environmental simulation equipment, it outputs a full-range standard signal, collects raw data for eight parameters, and performs fitting calculations for four different models. Based on the fitting results, the optimal calibration model is automatically selected: for pressure and differential pressure with excellent linearity, a linear calibration model is chosen; for temperature, relative humidity, dew point, and absolute humidity with significant nonlinear errors, a second-order polynomial calibration model is matched; for oxygen concentration exhibiting slight zero-point drift, offset calibration is applied; and for particle number detection with a fixed gain bias, proportional calibration is enabled. Simultaneously, uniform calibration of all acquisition channels is completed to eliminate acquisition biases caused by hardware channel differences.

[0121] Step 2: Multi-parameter cross-coupling error compensation. During normal operation, real-time measured data of temperature, pressure, relative humidity, oxygen concentration, and particle count are extracted to construct a dynamic five-parameter feature vector. A pre-trained cross-compensation matrix adapted to the cleanroom conditions is called to calculate the coupling compensation amount of each parameter in real time. This compensates for the interference of temperature and humidity fluctuations on dew point and absolute humidity, as well as the impact of pressure fluctuations on particle count detection accuracy, thus completing the multi-parameter coupling error correction and outputting stable intermediate calibration data.

[0122] Step 3: RLS Online Iteration and Stability Control. After the equipment enters normal monitoring mode, it continuously collects on-site time-series data. Using the true values ​​of standard environmental parameters in the cleanroom as a reference, the calibration weights are iteratively optimized frame by frame using the RLS algorithm. Corresponding fluctuation thresholds and convergence thresholds are set. When the data changes instantaneously due to the opening and closing of workshop doors, personnel movement, or airflow disturbances, the weight update is automatically paused to avoid calibration parameter errors. When the data fluctuation is within a reasonable range for more than 10 consecutive minutes and the weights tend to stabilize and converge, the latest calibration parameters are automatically fixed, completing the adaptive optimization update.

[0123] Step 4: Continuous closed-loop calibration throughout the entire process. The equipment continuously executes a three-layer calibration logic throughout its operation, correcting accuracy errors caused by sensor aging drift and gradual changes in operating conditions in real time, and continuously ensuring the accuracy and stability of multi-parameter acquisition.

[0124] 1.3 Implementation Results After 30 days of continuous on-site trial operation, the optimization effect of this proposed solution is significant. The accuracy of the equipment's eight environmental parameter acquisitions has been greatly improved, with temperature error controlled within ±0.1℃, humidity error ≤ ±0.8%RH, pressure and differential pressure error ≤ ±0.15%FS, oxygen concentration error ≤ ±0.3%VOL, and particle number and dew point parameter errors reduced by over 60%. Multi-parameter coupling interference has been largely eliminated, the consistency of synchronously acquired data has been greatly improved, and the cumulative error of long-term operation is less than 1.5%. The entire system requires no manual shutdown for calibration, has strong automated operation capabilities, and fully meets the stringent requirements of biopharmaceutical cleanrooms for high-precision, high-stability, and high-continuity environmental monitoring.

[0125] Combination Figure 3 This application also provides a calibration device for an environmental parameter acquisition unit applicable to the aforementioned electronic device. The calibration device for the environmental parameter acquisition unit may include a measured value vector construction module, a first calibration processing module, a second calibration processing module, a third calibration processing module, and a calibration data determination module.

[0126] The measured value vector construction module is used to acquire multiple raw environmental parameter data of various types collected by the environmental parameter collector at a target time, and to construct a target measured value vector based on the multiple raw environmental parameter data. In this embodiment, the measured value vector construction module can be used to perform... Figure 2 For details regarding step S110, the relevant content of the measured value vector construction module can be found in the preceding description of step S110.

[0127] The first calibration processing module is used to perform a first calibration process on the target measured value vector based on a pre-configured first standard value vector to obtain a first calibration vector. The first standard value vector includes multiple standard environmental parameter data corresponding to the various types collected over historical time, and the first calibration vector includes multiple first calibration environmental parameter data corresponding to the various types. In this embodiment, the first calibration processing module can be used to execute... Figure 2 The relevant content regarding the first calibration processing module in step S120 shown can be found in the previous description of step S120.

[0128] The second calibration processing module is used to perform a second calibration process on the first calibration vector based on the coupling relationship between the multiple types, to obtain a second calibration vector, wherein the second calibration vector includes multiple second calibration environment parameter data corresponding to the multiple types. In this embodiment, the second calibration processing module can be used to execute... Figure 2 For details regarding step S130 shown, please refer to the preceding description of step S130 for information about the second calibration processing module.

[0129] The third calibration processing module is used to perform a third calibration process on the second calibration vector based on a pre-determined fixed mapping calibration matrix to obtain a third calibration vector. The fixed mapping calibration matrix consists of parameters iterated and fixed using a recursive least squares algorithm based on a pre-configured second standard value vector. The second standard value vector includes multiple standard environmental parameter data corresponding to the various types collected over historical time. The third calibration vector includes multiple third calibration environmental parameter data corresponding to the various types. In this embodiment, the third calibration processing module can be used to execute... Figure 2 The relevant content regarding the third calibration processing module in step S140 shown can be found in the previous description of step S140.

[0130] The calibration data determination module is used to obtain multiple target calibration environment parameter data corresponding to the multiple original environment parameter data based on the third calibration vector, wherein there is a one-to-one correspondence between the multiple target calibration environment parameter data and the multiple original environment parameter data. In this embodiment, the calibration data determination module can be used to perform... Figure 2 The relevant content regarding the calibration data determination module in step S150 shown can be found in the previous description of step S150.

[0131] In this embodiment of the application, corresponding to the above-described calibration method for an environmental parameter acquisition device applied to the electronic device, a computer-readable storage medium is also provided. This computer-readable storage medium stores a computer program that, when executed, performs each step of the calibration method for the environmental parameter acquisition device. The steps executed by the aforementioned computer program are not described in detail here, but can be found in the preceding explanation of the calibration method for the environmental parameter acquisition device.

[0132] In summary, the calibration method, apparatus, device, and medium for the environmental parameter acquisition device provided in this application firstly acquire multiple raw environmental parameter data corresponding to various types collected at a target time, and construct a target measured value vector based on the multiple raw environmental parameter data; secondly, perform a first calibration process on the target measured value vector based on a first standard value vector to obtain a first calibration vector; then, perform a second calibration process on the first calibration vector based on the coupling correlation between various types to obtain a second calibration vector; further, perform a third calibration process on the second calibration vector based on a fixed mapping calibration matrix to obtain a third calibration vector; finally, obtain multiple target calibration environmental parameter data based on the third calibration vector. Based on the above, on the one hand, the first calibration process enables calibration based on standard environmental parameter data, which can reduce acquisition deviations caused by inherent sensor errors and hardware channel differences to a certain extent. On the other hand, the second calibration process enables calibration based on coupling correlation, which can offset interference errors caused by dynamic environmental changes to a certain extent, effectively improving the data consistency of multi-parameter synchronous acquisition. Furthermore, the third calibration process enables calibration based on a fixed mapping calibration matrix, which can improve the long-term accuracy drift caused by sensor aging and gradual changes in operating conditions to a certain extent, achieving dynamic calibration throughout the equipment's lifecycle. Therefore, by combining these three calibration processes, comprehensive data calibration can be achieved, obtaining reliable target calibration environmental parameter data, thereby improving the problem of low reliability in parameter calibration in existing technologies.

[0133] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus and method embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0134] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0135] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, 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, electronic device, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks. It should be noted that, in this document, 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. In the absence of further restrictions, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0136] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A calibration method for an environmental parameter collector, characterized in that include: Acquire multiple raw environmental parameter data of various types collected by the environmental parameter collector at the target time, and construct a target measured value vector based on the multiple raw environmental parameter data; Based on a pre-configured first standard value vector, the target measured value vector is subjected to a first calibration process to obtain a first calibration vector. The first standard value vector includes multiple standard environmental parameter data corresponding to the multiple types collected in historical time, and the first calibration vector includes multiple first calibration environmental parameter data corresponding to the multiple types. Based on the coupling relationship between the various types, the first calibration vector is subjected to a second calibration process to obtain a second calibration vector, wherein the second calibration vector includes multiple second calibration environment parameter data corresponding to the various types; Based on a predetermined solidified mapping calibration matrix, the second calibration vector is subjected to a third calibration process to obtain a third calibration vector. The solidified mapping calibration matrix is ​​a parameter that is iterated and solidified based on a pre-configured second standard value vector through a recursive least squares algorithm. The second standard value vector includes multiple standard environmental parameter data corresponding to the various types collected in historical time. The third calibration vector includes multiple third calibration environmental parameter data corresponding to the various types. Based on the third calibration vector, multiple target calibration environment parameter data corresponding to the multiple original environment parameter data are obtained, wherein there is a one-to-one correspondence between the multiple target calibration environment parameter data and the multiple original environment parameter data.

2. The calibration method of an environmental parameter collector according to claim 1, characterized in that, The step of performing a first calibration process on the target measured value vector based on a pre-configured first standard value vector to obtain a first calibration vector includes: Based on a pre-configured first standard value vector and historical measured value vector, each type of calibration model is fitted and determined to obtain a target calibration model corresponding to each type. The historical measured value vector includes multiple historical environmental parameter data corresponding to the various types, which are collected by the environmental parameter collector at the historical time corresponding to the first standard value vector. The target calibration model includes an offset calibration model, a proportional calibration model, a linear calibration model, or a quadratic polynomial calibration model, and the fitting and determination are based on the coefficient of determination and / or the mean absolute error. Based on the target calibration model corresponding to each type, the original environmental parameter data of the corresponding type in the target measured value vector are subjected to the first calibration process to obtain the first calibration vector.

3. The calibration method of an environmental parameter collector according to claim 1, characterized in that, The step of performing a second calibration process on the first calibration vector based on the coupling relationship between the various types to obtain a second calibration vector includes: Based on the multiple historical environmental parameter data and constant terms corresponding to the various types collected by the environmental parameter collector in historical time, a target input vector is constructed. The target input vector is solved by batch least squares to obtain the target weight matrix and the target bias vector, wherein the target weight matrix and the target bias vector are used to characterize the coupling relationship between the various types; Based on the target weight matrix and the target bias vector, the first calibration vector is subjected to a second calibration process to obtain a second calibration vector.

4. The calibration method for the environmental parameter acquisition device according to claim 1, characterized in that, The step of performing a third calibration process on the second calibration vector based on a pre-determined fixed mapping calibration matrix to obtain a third calibration vector includes: Based on a predetermined fixed mapping calibration matrix, the second calibration vector is subjected to a third calibration process to obtain a third calibration result, wherein the third calibration result includes multiple third calibration environment parameter data corresponding to the various types; When the third calibration result is within the range of the environmental parameter acquisition device, the third calibration result is determined as the third calibration vector; When the result of the third calibration process is not within the range of the environmental parameter acquisition device, a new fixed mapping calibration matrix is ​​obtained by iterating and solidifying the result using a recursive least squares algorithm.

5. The calibration method for the environmental parameter acquisition device according to any one of claims 1-4, characterized in that, The calibration method for the environmental parameter acquisition device further includes the step of forming the solidified mapping calibration matrix, which includes: Based on multiple second standard value vectors corresponding to multiple historical times, an initial matrix is ​​constructed, forming the initial coefficient vector and the initial covariance inverse matrix of the recursive least squares algorithm; Based on the environmental parameter data collected by the environmental parameter collector corresponding to the last historical time among the multiple historical times, and the output data after the first and second calibration processes, the current iteration input vector is constructed. Based on the second standard value vector corresponding to the last historical time among the multiple historical times, anomaly filtering judgment is performed on the current iteration input vector. When the current iteration input vector does not need to be filtered, the initial coefficient vector and the initial covariance inverse matrix are updated based on the current iteration input vector and the initial matrix to obtain the updated coefficient vector and the updated covariance inverse matrix. The updated coefficient vector and the updated covariance inverse matrix are used as the objects of the next update. Based on the updated coefficient vector, the solidification mapping calibration matrix is ​​determined.

6. The calibration method for the environmental parameter acquisition device according to claim 5, characterized in that, The step of determining the solidified mapping calibration matrix based on the updated coefficient vector includes: Based on the updated coefficient vector, the current iteration input vector is mapped, and the mapping result is adjusted to the second standard value vector corresponding to the last historical time in the plurality of historical times according to a preset convergence strength, so as to obtain the current iteration output vector. The solidification is determined based on the current iteration output vector, and when it is determined that solidification is possible, the updated coefficient vector is determined as the solidification mapping calibration matrix.

7. The calibration method for the environmental parameter acquisition device according to claim 6, characterized in that, The step of determining the solidification based on the current iteration output vector, and determining the updated coefficient vector as the solidification mapping calibration matrix when it is determined that solidification is possible, includes: At least the current iteration output vector is subjected to range constraint judgment and rate of change constraint judgment; When the current iterative output vector satisfies the range constraint judgment condition and the rate of change constraint judgment condition, the current iterative output vector is determined as a stable output vector, and the current stable duration is obtained. When the current iterative output vector does not satisfy the range constraint judgment condition or the rate of change constraint judgment condition, the current iterative output vector is not determined as a stable output vector, and the current stable duration is cleared. The stable duration is used to characterize the duration from the first determination of the stable output vector after the clearing process to the current time. Based on the current stable duration, determine whether the preset curing conditions are met, and if the preset curing conditions are met, determine the updated coefficient vector as the curing mapping calibration matrix.

8. A calibration device for an environmental parameter acquisition device, characterized in that, include: The measured value vector construction module is used to acquire multiple raw environmental parameter data of various types collected by the environmental parameter collector at the target time, and to construct a target measured value vector based on the multiple raw environmental parameter data. The first calibration processing module is used to perform a first calibration process on the target measured value vector based on a pre-configured first standard value vector to obtain a first calibration vector. The first standard value vector includes multiple standard environmental parameter data corresponding to the multiple types collected in historical time, and the first calibration vector includes multiple first calibration environmental parameter data corresponding to the multiple types. The second calibration processing module is used to perform a second calibration processing on the first calibration vector based on the coupling relationship between the multiple types to obtain a second calibration vector, wherein the second calibration vector includes multiple second calibration environment parameter data corresponding to the multiple types; The third calibration processing module is used to perform a third calibration process on the second calibration vector based on a pre-determined solidified mapping calibration matrix to obtain a third calibration vector. The solidified mapping calibration matrix is ​​a parameter that is iterated and solidified based on a pre-configured second standard value vector through a recursive least squares algorithm. The second standard value vector includes multiple standard environmental parameter data corresponding to the various types collected in historical time. The third calibration vector includes multiple third calibration environmental parameter data corresponding to the various types. The calibration data determination module is used to obtain multiple target calibration environment parameter data corresponding to the multiple original environment parameter data based on the third calibration vector, wherein there is a one-to-one correspondence between the multiple target calibration environment parameter data and the multiple original environment parameter data.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor connected to the memory is used to execute the computer program stored in the memory to implement the calibration method of the environmental parameter acquisition device according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains a computer program that, when executed, performs the calibration method for the environmental parameter acquisition device according to any one of claims 1-7.

Images