Control method and device for environment control equipment of processing workshop, equipment and medium

Abstract

The invention discloses a control method and device for environment control equipment of a processing workshop, equipment and a medium. The control method comprises the steps that at least one equipment signal sent by the environment control equipment is received; constructing a nonlinear model according to the at least one equipment signal, and determining a system category of the environment control equipment according to the nonlinear model; if the category of the environment control equipment is a weak non-linear category, performing weak non-linear processing on the at least one equipment signal to obtain first intervention data; if the category of the environment control equipment is a strong nonlinear category, performing strong nonlinear processing on the at least one equipment signal to obtain second intervention data; and sending the first intervention data or the second intervention data to the environment control equipment. According to the method, the problem that the computing power and the resource consumption of the controller are too large due to the fact that the environment control equipment is controlled by adopting a nonlinear model at present is avoided, and the control precision and the reliability of the environment control equipment are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and particularly to a control method, device, equipment and medium for an environmental control device in a processing workshop. Background Art

[0002] Currently, a controller is usually connected to multiple environmental control devices in a processing workshop to control the output data of each environmental control device; and each processing device in the processing workshop will respectively change its working environment through at least one environmental control device to make the working state of each processing device reach the best.

[0003] The controller currently used to control environmental control devices usually uses a linear model for control management. However, the traditional linear model is difficult to describe the sudden change in the behavior of the device near the critical point (for example: phase change of materials caused by sudden change in temperature), resulting in low control accuracy and reliability of the environmental control device. And if a non-linear model is called for all environmental control devices for control, it will greatly consume the computing power and resources of the controller. Summary of the Invention

[0004] The purpose of the present invention is to provide a control method, device, equipment and medium for an environmental control device in a processing workshop, which is used to solve the problems in the prior art that lead to low control accuracy and reliability of the environmental control device, and will greatly consume the computing power and resources of the controller.

[0005] To achieve the above purpose, the present invention provides a control method for an environmental control device, including: Receiving at least one device signal sent by an environmental control device; wherein, the device signal includes an input data and an output data; the input data is used to characterize the environmental condition of the space where the environmental control device is located or the operating condition of the environmental control device at a time point; the output data is the control data for the environmental control device to change or maintain the environmental condition or the operating condition at a time point; Constructing a non-linear model according to at least one of the device signals, and determining the system category of the environmental control device according to the non-linear model; wherein, the non-linear model reflects the non-linear degree of the environmental control device; the system category includes weak non-linearity and strong non-linearity; If the category of the environmental control device is a weak non-linear category, performing weak non-linear processing on at least one of the device signals to obtain first intervention data; the weak non-linear processing is one of linearization processing, PID control processing and state feedback control processing; the first intervention data is used as the output data of the environmental control device to make the input data reach a preset target index; If the category of the environmental control device is a strongly non-linear category, perform strongly non-linear processing on at least one of the device signals to obtain second intervention data; wherein, the strongly non-linear processing is adaptive control processing or fuzzy control processing; the second intervention data is used as the output data of the environmental control device to make the input data reach a preset target index; Send the first intervention data or the second intervention data to the environmental control device.

[0006] In the above solution, receiving at least one device signal sent by an environmental control device includes: Send an interaction instruction to an environmental control device; wherein, the interaction instruction has any one of a time signal, a time period signal, and a quantity signal; the time signal has a time value, and the time signal is used to instruct the environmental control device to send the device signal within the time value; the time period signal has at least one time period, and the time period signal is used to instruct the environmental control device to send the device signal within the time period; the quantity signal has a quantity value, and the quantity signal is used to instruct the environmental control device to send the device signal of the quantity value; Receive at least one device signal sent by the environmental control device according to the interaction instruction.

[0007] In the above solution, constructing a non-linear model according to at least one of the device signals includes: Construct an analysis model of the environmental control device according to at least one of the device signals; Determine the lag order in the analysis model, and select a basis function according to the system characteristic information; Based on the analysis model, construct the non-linear model of the environmental control device according to the lag order and the basis function.

[0008] In the above solution, determining the system category of the environmental control device according to the non-linear model includes: Combine at least one basis function in the non-linear model into a regression matrix; Convert the non-linear model into a linear function according to the regression matrix; Determine the non-linear contribution rate of the environmental control device according to the linear function; wherein, the non-linear contribution rate characterizes the importance of the non-linear term in the non-linear model in the non-linear model Determine the system category of the environmental control device according to the non-linear contribution rate.

[0009] In the above solution, performing weakly non-linear processing on at least one of the device signals to obtain first intervention data includes: If it is determined that the environmental control device is of the first type, linearize at least one of the device signals to obtain first intervention data; wherein, the first type indicates that the output data of the environmental control device has one output variable, and by adjusting the output data of the environmental control device within a first range near a working point, the input data of the environmental control device reaches a preset first input index; If it is determined that the environmental control device is of the second type, perform PID control processing on at least one of the device signals to obtain first intervention data; wherein, the second type indicates that the output data of the environmental control device has one output variable, and it is necessary to adjust the output data of the environmental control device within a preset second time so that the input data of the environmental control device reaches a preset second input index; If it is determined that the environmental control is of the third type, perform state feedback control processing on at least one of the device signals to obtain first intervention data; wherein, the third type indicates that the output data of the environmental control device has two or more output variables, and it is necessary to adjust at least one of the output variables so that the input data of the environmental control device reaches a preset third input index.

[0010] In the above solution, performing strong non-linear processing on at least one of the device signals to obtain second intervention data includes: If it is determined that the environmental control device is of the fourth type, perform adaptive control processing on at least one of the device signals to obtain second intervention data; wherein, the fourth type indicates that the device signal generated by the environmental control device can be linearly processed; If it is determined that the environmental control device is of the fifth type, perform fuzzy control processing on at least one of the device signals to obtain second intervention data; wherein, the fifth type indicates that the device signal generated by the environmental control device cannot establish a mathematical model.

[0011] In the above solution, sending the first intervention data or the second intervention data to the environmental control device as the output data of the environmental control device includes: Generating intervention indication information according to the first intervention data or the second intervention data; wherein, the intervention indication information is used to instruct the environmental control device to use the first intervention data or the second intervention data as the output data; Sending the intervention indication information to the environmental control device.

[0012] To achieve the above object, the present invention also provides a control device for an environmental control device, including: An input module for receiving at least one device signal sent by an environmental control device; wherein the device signal includes an input data and an output data; the input data is used to characterize the environmental condition of the space where the environmental control device is located or the operating condition of the environmental control device at a time point; the output data is control data for the environmental control device to change or maintain the environmental condition or the operating condition at a time point; A category recognition module for constructing a non - linear model based on at least one of the device signals and determining the system category of the environmental control device according to the non - linear model; wherein the non - linear model reflects the non - linear degree of the environmental control device; the system category includes weak non - linearity and strong non - linearity; A weak non - linear module for performing weak non - linear processing on at least one of the device signals to obtain first intervention data if the category of the environmental control device is a weak non - linear category; the weak non - linear processing is one of linearization processing, PID control processing, and state - feedback control processing; the first intervention data is used as the output data of the environmental control device to make the input data reach a preset target index; A strong non - linear module for performing strong non - linear processing on at least one of the device signals to obtain second intervention data if the category of the environmental control device is a strong non - linear category; wherein the strong non - linear processing is adaptive control processing or fuzzy control processing; the second intervention data is used as the output data of the environmental control device to make the input data reach a preset target index; An output module for sending the first intervention data or the second intervention data to the environmental control device.

[0013] To achieve the above object, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor of the electronic device executes the computer program, the steps of the control method of the above - mentioned environmental control device are implemented.

[0014] To achieve the above object, the present invention also provides a computer - readable storage medium, on which a computer program is stored. When the computer program stored on the readable storage medium is executed by a processor, the steps of the control method of the above - mentioned environmental control device are implemented.

[0015] A control method, device, equipment, and medium for an environmental control device in a processing workshop provided by the present invention, by proposing device signals with input data and output data composed of time series, analyzing the time series of the device signals, automatically identifying the non - linear degree of the system, and avoiding the subjectivity and trial - and - error cost of manual modeling.

[0016] By constructing a non - linear model based on at least one of the device signals and determining the system category of the environmental control device according to the non - linear model, accurate classification of the system classification of the environmental control device is achieved, solving the problem of mis - selection of control strategies caused by traditional classification methods, and avoiding the problem of excessive computing power and resource consumption of the controller caused by the current use of non - linear models for all environmental control devices.

[0017] By linearizing at least one of the device signals to obtain the first intervention data, the non - linear equation constructed by at least one device signal is transformed into a linear model, and the first intervention data is generated based on this linear model, achieving the technical effect of quickly adjusting the output data of the environmental control device to make the input data reach the preset target index, ensuring the control accuracy and reliability of the environmental control device, and solving the problem of gain drift caused by component aging.

[0018] By performing PID control processing on at least one of the device signals to obtain the first intervention data, it is possible to calculate in real - time the error generated during the operation of the environmental control device according to the device signal, accurately adjust the output data of the environmental control device, and achieve the technical effect of making the input data reach the preset target index, ensuring the control accuracy and reliability of the environmental control device.

[0019] By performing state - feedback control processing on at least one of the device signals to obtain the first intervention data, the adjustment of the output data of a multi - variable system is realized. The multi - variable system has input data from multiple input units, and one input unit represents an environmental condition or operating state, achieving the technical effect of making the input data reach multiple preset target indexes respectively, ensuring the control accuracy and reliability of the environmental control device, and expanding the application range.

[0020] By performing adaptive control processing on at least one of the device signals to obtain the second intervention data, the output data of the environmental control device that can establish a non - linear mathematical model generated by the device signal is adjusted in real - time to ensure the control accuracy and reliability of the environmental control device.

[0021] By performing fuzzy control processing on at least one of the device signals to obtain the second intervention data, the output data of the environmental control device that cannot establish a mathematical model is adjusted to make the input data reach the preset target index, ensuring the control accuracy and reliability of the environmental control device, and expanding the application range. Brief Description of the Drawings

[0022] Figure 1 It is a flowchart of the first embodiment of the control method for the environmental control device of the present invention; Figure 2Schematic diagram of the program module of the second embodiment of the control method and device of the environmental control device of the present invention; Figure 3 Schematic diagram of the hardware structure of the electronic device in the third embodiment of the electronic device of the present invention. Detailed implementation manners

[0023] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0024] Embodiment 1: Please refer to Figure 1 , a control method for an environmental control device in a processing workshop according to this embodiment includes: S101: Receive at least one device signal sent by an environmental control device; wherein, the device signal includes an input data and an output data; the input data is used to characterize the environmental condition of the space where the environmental control device is located or the operating condition of the environmental control device at a time point; the output data is the control data for the environmental control device to change or maintain the environmental condition or the operating condition at a time point; S102: Construct a non-linear model according to at least one of the device signals, and determine the system category of the environmental control device according to the non-linear model; wherein, the non-linear model reflects the non-linear degree of the environmental control device; the system category includes weak non-linearity and strong non-linearity; S103: If the category of the environmental control device is a weak non-linear category, perform weak non-linear processing on at least one of the device signals to obtain first intervention data; the weak non-linear processing is one of linearization processing, PID control processing and state feedback control processing; the first intervention data is used as the output data of the environmental control device to make the input data reach a preset target index; S104: If the category of the environmental control device is a strong non-linear category, perform strong non-linear processing on at least one of the device signals to obtain second intervention data; wherein, the strong non-linear processing is adaptive control processing or fuzzy control processing; the second intervention data is used as the output data of the environmental control device to make the input data reach a preset target index; S105: Send the first intervention data or the second intervention data to the environmental control device.

[0025] In this example, it is difficult for traditional linear models to describe the behavioral mutations of devices near critical points (such as temperature mutations leading to material phase transitions). By constructing non-linear models (such as Volterra series, NARMAX models), the non-linear mapping relationship between input and output can be captured. For example, in the temperature control of a chemical reactor, the activity of the catalyst is non-linearly correlated with temperature, and a linear model will result in a control error of ±15%.

[0026] In response, this application proposes device signals with input data and output data composed based on time series. Through time series analysis of the device signals, the non-linear degree of the system is automatically identified, avoiding the subjectivity of manual modeling and the trial-and-error cost.

[0027] Secondly, a non-linear model is constructed based on at least one of the device signals, and the system category of the environmental control device is determined according to the non-linear model, achieving accurate classification of the system classification of the environmental control device, and solving the problem of misselection of control strategies caused by traditional classification methods, such as misjudging a weakly non-linear system as a strongly non-linear one and over-controlling, increasing energy consumption.

[0028] Thirdly, by linearizing at least one of the device signals to obtain the first intervention data, the non-linear equation constructed by at least one device signal is transformed into a linear model by ignoring high-order infinitesimal terms through Taylor expansion, and the first intervention data is generated based on this linear model, achieving rapid adjustment of the output data of the environmental control device to make the input data reach the preset target index, ensuring the control accuracy and reliability of the environmental control device, and solving the problem of gain drift caused by component aging. For example: The first intervention data is used as a compensation signal to correct the temperature set value, enabling the actual output data of the environmental control device (such as the temperature control system of a processing device or precision instrument) to help the processing device served by the environmental control device to reach the predetermined target index of the processing device more quickly and accurately, that is, the input data of the environmental control device reaches the target index at this time. In this embodiment, the target index is the output data of the processing device, and the output data of the processing device includes: temperature, humidity, flow rate, rotational speed, current, voltage, etc.

[0029] By performing PID control processing on at least one of the device signals to obtain the first intervention data, the error generated during the operation of the environmental control device is calculated in real time according to the device signals, and the output data of the environmental control device is accurately adjusted to achieve the technical effect of making the input data reach the preset target index, ensuring the control accuracy and reliability of the environmental control device. For example: In chemical fluid control, when the pipeline pressure fluctuates due to changes in the viscosity of the medium, the first intervention data dynamically adjusts the differential coefficient in the environmental control device to suppress the overshoot, and then enables the flow rate of the processing device served by the environmental control device to reach the predetermined target index.

[0030] By performing state feedback control processing on at least one of the device signals to obtain first intervention data, the adjustment of the output data of the multivariable system is realized. The multivariable system operates with the input data of multiple input units, and one input unit represents an environmental condition or an operating state, so that the input data respectively reaches multiple preset target indicators, ensuring the control accuracy and reliability of the environmental control device and expanding the application range.

[0031] Finally, by performing adaptive control processing on at least one of the device signals to obtain second intervention data, the generated device signal can perform real-time adjustment on the output data of the environmental control device that can establish a non-linear mathematical model, so as to ensure the control accuracy and reliability of the environmental control device, and further eliminate the cross influence between the attitude angle and the position control. For example: for a parameter time-varying system (such as the attitude control of an unmanned aerial vehicle in gusts), the second intervention data updates the control law online based on the Lyapunov stability theory, enabling the system to have an adaptive ability to changes in mass distribution.

[0032] By performing fuzzy control processing on at least one of the device signals to obtain second intervention data, the output data of the environmental control device that cannot establish a mathematical model is adjusted, so that the input data reaches the preset target indicators, ensuring the control accuracy and reliability of the environmental control device. For example: in a system that cannot establish a mathematical model (such as fermentation process control), the second intervention data realizes the embedding of expert knowledge through a fuzzy rule base (such as "higher temperature → reduce heating power").

[0033] In this embodiment, HarmonyOS is installed in the controller that runs the control method of the environmental control device of the processing workshop and in the environmental control device respectively. By installing HarmonyOS in the controller, the controller can obtain control signals from the environmental control device without encoding and decoding operations, and send the first intervention data and the second intervention data to the environmental control device, realizing unobstructed interactive communication between the controller and each environmental control device.

[0034] In a preferred embodiment, receiving at least one device signal sent by an environmental control device includes: Send an interaction instruction to an environmental control device; among them, any one of a time signal, a time period signal, and a quantity signal is included in the interaction instruction; a time value is included in the time signal, and the time signal is used to instruct the environmental control device to send a device signal within the time value; at least one time period is included in the time period signal, and the time period signal is used to instruct the environmental control device to send a device signal within the time period; a quantity value is included in the quantity signal, and the quantity signal is used to instruct the environmental control device to send a device signal of this quantity value; Receive at least one device signal sent by the environmental control device according to the interaction instruction.

[0035] In this example, environmental control devices often cause disordered data transmission due to the lack of a time reference (such as out-of-order reporting of Internet of Things sensors). Real-time systems require microsecond-level synchronization (such as industrial PLC control) but are difficult to achieve. In response to this, the time signal provides an absolute time anchor (such as GPS time synchronization + local crystal oscillator calibration), enabling the synchronization accuracy between devices to reach ±1 μs (compared with ±10 ms without a time signal); the pulse-triggered mechanism supports zero-delay response (such as 5G TTI scheduling), solving the inherent delay of the traditional polling mechanism.

[0036] Concurrent transmission of multiple devices is likely to cause network congestion (such as LoRaWAN gateway conflicts); static time slot allocation results in low channel utilization (such as fixed allocation of NB-IoT subcarriers).

[0037] In response to this, the time period signal implements dynamic time division multiple access (D-TDMA), reducing the channel conflict rate; the adaptive time period compression algorithm dynamically adjusts the time period length according to network load, improving the spectrum efficiency by 300%.

[0038] Sudden traffic is likely to cause buffer overflows (such as missed reports in the in-vehicle ETSI-ITS-G5 standard); fixed sampling rates generate redundant data (such as AMI meter reading in smart grids). In response to this, the quantity signal cooperates with the token bucket algorithm to achieve stepped traffic shaping; the abnormal traffic fusing mechanism automatically triggers ACK suspension when the quantity value exceeds the standard (such as MQTT protocol optimization), avoiding network avalanches.

[0039] The clock accuracies of different devices vary greatly (such as ±50 ppm for consumer-grade devices vs. ±0.1 ppm for industrial-grade devices); differences in protocol stacks cause control instructions to fail (such as in scenarios with a mixture of CoAP / MQTT / HTTP).

[0040] In this regard, the three-layer signal nested design (time - time period - quantity) realizes a cross - protocol translation layer, supporting transparent conversion from Modbus / TCP to MQTT; the adaptive compensation algorithm automatically corrects the time signal according to the device clock drift (such as Kalman filter calibration), with the accuracy improved by two orders of magnitude.

[0041] Exemplarily, the instruction is sent to the target environmental control device through a communication module (such as an MQTT Topic or an HTTP request). The instruction format can use JSON, Protobuf, or binary encapsulation.

[0042] If the interaction instruction has a time signal, the interaction instruction is used to start the timer of the environmental control device and instruct the environmental control device to send device signals at a fixed frequency (for example, 1Hz) within a specified duration.

[0043] If the interaction instruction has a time - period signal, the interaction instruction is used to instruct the environmental control device to enter the timing waiting logic and automatically activate the environmental control device to send the device signal within a specified time period.

[0044] If the interaction instruction has a quantity signal, the interaction instruction is used to instruct the environmental control device to call a counter, and the count is decremented by one after each device signal is sent until it reaches zero and stops.

[0045] The environmental control device returns the device signal through the same communication channel. The master control end listens to the response channel and processes, verifies, and records the device signal.

[0046] Refer to the NTP protocol to synchronize the clock of the environmental control device with the clock of the controller running the control method of the environmental control device.

[0047] The interaction instruction has a unique identifier ID, a verification field, and an expiration time field to ensure data integrity.

[0048] Control the sending frequency to avoid network congestion or excessive device load.

[0049] Adapt to multiple protocol communication formats (MQTT / CoAP / HTTP) and can be deployed in embedded or cloud systems.

[0050] In a preferred embodiment, a non - linear model is constructed based on at least one of the device signals, including: Construct an analysis model of the environmental control device based on at least one of the device signals; Determine the lag order in the analysis model and select a basis function according to the system characteristic information; Based on the analysis model, construct the non - linear model of the environmental control device according to the lag order and the basis function.

[0051] In this example, the lag order is selected by the AIC / BIC criterion to match the model complexity with the amount of data (to avoid overfitting / underfitting); multi - order lag terms are used to capture non - stationary characteristics such as quadratic phase coupling (such as in radio frequency power amplifier modeling). The basis function is selected according to the system characteristic information, so that the obtained non - linear model has enhanced sparsity, improved physical interpretability, optimized numerical stability, enhanced noise resistance and improved computational efficiency.

[0052] Exemplarily, the general form of the analysis model is: y(t)=F[y(t - 1),…,y(t - n),u(t - 1),…,u(t - m)]+e(t).

[0053] Where, y(t): is the output data at time point t in the device signal; y(t - 1): is the output data at time point (t - 1) in the device signal; u(t): is the input data in the device signal; F[]: is the basis function, and the basis function is any one of polynomial function, Gaussian function, multi - quadratic function, trigonometric function, exponential function, Fourier series function, sine function, cosine function, orthogonal polynomial; n: the lag order of the output data; m: is the lag order of the input data; e(t): modeling error or noise; Determine the lag order n of the output data and the lag function m of the input data through the autocorrelation function (ACF), cross - correlation function (CCF) and partial autocorrelation function (PACF); or Determine the lag order n of the output data and the lag function m of the input data through the AIC / BIC criterion.

[0054] Specifically, determining the lag order n of the output data and the lag function m of the input data through the autocorrelation function (ACF) and cross - correlation function (CCF) includes: Use the statsmodels library to plot the autocorrelation function graph; where the autocorrelation function graph is used to show the correlation between the time series and its own lags at each order, and is used to identify the periodicity or memory of the sequence; in this embodiment, the vertical axis of the autocorrelation function graph is the value of the output data at each time point, and the horizontal axis is the lag value of the output data with respect to its own lags at each order; When the output line in the autocorrelation function graph first falls within the confidence interval, the value of this output line on the horizontal axis is used as the lag order of the output data. For example: Suppose the ACF graph first falls within the confidence interval (such as becoming significantly 0 for the first time) after lag k = 2, then n = 2 is selected.

[0055] Use the statsmodels library to plot the cross-correlation graph; among them, it is a tool used in time series analysis to measure the linear relationship between two different time series. It shows the correlation coefficients of two time series at different lag orders and helps identify the leading-lagging relationship between the two series; in this embodiment, The cross-correlation graph reflects the degree of linear correlation between the input time series and the output time series at at least one lag order and obtains the correlation coefficient; through the cross-correlation graph, the leading-lagging relationship between the two time series can be identified. For example, in time series prediction, the cross-correlation between the input data and the output data can be analyzed to determine the lag effect of the input data on the output data. The input time series reflects at least one input data and the lag value of each said input data at one order; the output time series reflects at least one output data and the lag value of each said output data at one order.

[0056] Take the lag order with the largest correlation coefficient as the lag order of the output data. For example: Suppose the CCF graph has the largest correlation coefficient at lag k = 1, then m = 1 is selected.

[0057] The system characteristic information of the environmental control device includes: low nonlinear systems, high nonlinear systems, systems with periodicity, and systems with specific boundary conditions.

[0058] A low nonlinear system refers to a system where the relationship between the output and the input is approximately linear, and the nonlinear effect can be ignored or is very weak. For example: The dimming control of a lighting system, where the lighting system controls the brightness of the light by adjusting the current. Within the normal operating range, the light output is approximately linearly related to the input current. Characteristics: The output (light output) and the input (current) are approximately linear. For low nonlinear systems: Simple basis functions can be selected, such as polynomial basis functions.

[0059] A high nonlinear system refers to a system where there is a significant nonlinear relationship between the output and the input, and a linear model cannot accurately describe the behavior of the system. For example: The temperature sensor in a temperature control system, which converts temperature changes into electrical signals, but temperature measurement involves complex nonlinear responses, including nonlinear characteristics such as the sensitivity and response time of the sensor. Characteristics: There is a significant nonlinear relationship between the output (electrical signal) and the input (temperature change). For highly nonlinear systems, polynomial basis functions or radial basis functions (RBFs), such as Gaussian functions and multiquadric functions, can be selected.

[0060] A system with periodicity means that the output or state of the system shows a periodic change pattern over time. For example, the timed opening and closing of smart curtains. Smart curtains can automatically open and close according to a preset schedule, and their opening and closing actions are periodic, such as opening in the morning and closing at night every day. Feature: The output (curtain opening and closing state) shows a periodic change over time. For a system with periodicity, Fourier basis functions such as sine and cosine functions can be selected.

[0061] A system with specific boundary conditions means that the output or state of the system satisfies specific constraint conditions at specific boundaries. For example, the access control of a smart door lock. The smart door lock controls access rights through methods such as passwords, fingerprints, or facial recognition, and the access rights are restricted by preset boundaries, such as only authorized users can unlock the door. Feature: The output (door lock state) is constrained at the preset boundary. For a system with specific boundary conditions, basis functions that meet these conditions can be selected, such as orthogonal polynomials.

[0062] Take the selected basis function as the basis function in the analysis model, and substitute the obtained lag order into the analysis model to obtain the nonlinear model of the environmental control device.

[0063] In a preferred embodiment, determining the system category of the environmental control device according to the nonlinear model includes: Combine at least one basis function in the nonlinear model into a regression matrix; Convert the nonlinear model into a linear function according to the regression matrix; Determine the nonlinear contribution rate of the environmental control device according to the linear function; wherein, the nonlinear contribution rate characterizes the importance of the nonlinear terms in the nonlinear model in the nonlinear model; Determine the system category of the environmental control device according to the nonlinear contribution rate.

[0064] In this example, traditional linear models cannot capture non-stationary characteristics such as quadratic phase coupling and chaotic motion (such as switching noise in power electronic converters). For this, through Volterra series expansion, project the nonlinear terms into the basis function space, construct a high-dimensional regression matrix, and reconstruct the unobservable internal state through the output signal (similar to the phase space reconstruction technique).

[0065] Directly calculating the Nth-order nonlinear term will result in an O(N²) parameter explosion (such as the curse of dimensionality in polynomial regression); for this, by using sparse coding techniques (such as LASSO regression), reduce the dimension of the basis function matrix to obtain a linear function, and on the premise of maintaining a 95% variance contribution rate, the number of parameters is reduced by 80% (IEEE benchmark test cases) There is a lack of a unified index to measure the impact of the non - linear term on the system output. In this regard, through Sobol variance decomposition, the output variance is decomposed into linear contribution and non - linear contribution to achieve the accurate calculation of the non - linear contribution rate (NCI).

[0066] Exemplarily, at least one basis function F[y(t - 1),…,y(t - n),u(t - 1),…,u(t - m)] in the non - linear model y(t)=F[y(t - 1),…,y(t - n),u(t - 1),…,u(t - m)]+e(t) is combined into a regression matrix Φ; according to the regression matrix Φ, the non - linear model is transformed into a linear function: y = Φθ+e(t), where θ is the parameter to be estimated, e(t) is the error term, and y is the output term of the linear function.

[0067] Specifically, determining the non - linear contribution rate of the environmental control device according to the linear function includes: Performing parameter estimation on the parameter to be estimated in the linear function to obtain a parameter estimation value; According to the parameter estimation value and the output data of the basis function in the linear function at each time point, obtaining the proportion of the unit non - linear effect contribution of the environmental control device at each time point; Calculating the mean value of the proportion of the unit non - linear effect contribution of the environmental control device at all time points to obtain the non - linear effect rate of the environmental control device.

[0068] Exemplarily, the parameter θ of the linear function is estimated by the least - squares method or the recursive least - squares method. Least - squares method (LS): Estimate the parameter θ by minimizing the sum of squared errors; Recursive least - squares method (RLS): Applicable to online parameter estimation, and update the parameter estimation value θ through a recursive algorithm.

[0069] The estimated parameters include a constant - term weight θ1 (for example: constant term), a linear - term weight θ2 (for example: the weight of the linear term y(t - 1)), and a non - linear - term weight θ3 (for example: the weight of the non - linear term y(t - 1)^2).

[0070] For example: the linear function is: y(t)=θ1+θ2y(t - 1)+θ3y(t - 1) 2 +e(t); Assume that the parameter estimation result is: θ = [0.5, 1.2, - 0.8]^T; The output of the basis function at time t: ϕ(t)=[1, y(t - 1), y(t - 1) 2 T; y(t - 1): is the output data at the (t - 1) time point in the device signal; then the proportion of non - linear contribution is: η=(|θ3|×|y(t - 1) 2|) / (|θ1| + |θ2| × |y(t - 1)| + |θ3| × |y(t - 1) 2 |) × 100% When y(t - 1) = 2: η = (0.8 × 4) / (0.5 + 1.2 × 2 + 0.8 × 4) × 100% ≈ 42.1%, which indicates that the contribution ratio of the unit non - linear effect of the environmental control device at time point t is approximately 42.1%.

[0071] Specifically, determining the system category of the environmental control device according to the non - linear contribution rate includes: If it is determined that the non - linear contribution rate does not exceed the preset non - linear threshold, then determine that the system category of the environmental control device is weakly non - linear; If it is determined that the non - linear contribution rate exceeds the non - linear threshold, then determine that the system category of the environmental control device is strongly non - linear.

[0072] Exemplarily, if the non - linear threshold is 50%; if the non - linear contribution rate of the environmental control device is 30%, then determine that the system category of the environmental control device is weakly non - linear; if the non - linear contribution rate of the environmental control device is 80%, then determine that the system category of the environmental control device is strongly non - linear.

[0073] In a preferred embodiment, performing weakly non - linear processing on at least one of the device signals to obtain first intervention data, including: If it is determined that the environmental control device is of the first type, then perform linearization processing on at least one of the device signals to obtain first intervention data; wherein, the first type characterizes that the output data of the environmental control device has one output variable, and by adjusting the output data of the environmental control device within a first range near a working point, the input data of the environmental control device reaches the preset first input index; If it is determined that the environmental control device is of the second type, then perform PID control processing on at least one of the device signals to obtain first intervention data; wherein, the second type characterizes that the output data of the environmental control device has one output variable, and within a preset second time, by adjusting the output data of the environmental control device, the input data of the environmental control device reaches the preset second input index; If it is determined that the environmental control is of the third type, then perform state - feedback control processing on at least one of the device signals to obtain first intervention data; wherein, the third type characterizes that the output data of the environmental control device has two or more output variables, and by adjusting at least one of the output variables, the input data of the environmental control device reaches the preset third input index.

[0074] In this example, the linearization process is based on Taylor expansion. By retaining the first-order term and neglecting the higher-order terms, the non-linear system is approximated as a linear system near a specific operating point. The system model is simplified by neglecting the higher-order terms, which facilitates the analysis and design of control strategies.

[0075] PID (Proportional-Integral-Derivative) control is a control algorithm based on the deviation. The control quantity is obtained through proportional, integral, and differential operations, and then the controlled object is controlled.

[0076] State feedback processing is a feedback method in which the state variables of the system are transmitted to the input end through a proportional link. Specifically, linearizing at least one of the control signals to obtain first intervention data includes: Constructing a non-linear model based on at least one of the control signals; Performing Taylor series expansion on at least one non-linear part in the non-linear model to obtain at least one expansion function; Respectively retaining the linear terms in at least one of the expansion formulas and respectively deleting the higher-order terms in at least one of the expansion formulas to obtain at least one linear function; Invoking the linear function to generate first intervention data based on the last input data.

[0077] Exemplarily, performing polynomial regression on at least one of the control signals to construct a polynomial regression model, for example: y(t)=β0 + β1u(t)+β2u(t) 2 +⋯+βnu(t) n ; where y(t) is the output data in the control signal; u(t) is the input data in the control signal, and β0, β1, β2, ⋯βn are the coefficients of each term in the initial non-linear model respectively; n is any natural number greater than 2; Using methods such as the least squares method, gradient descent method, or Bayesian optimization to minimize the prediction error: βmini = ∑(ytrue(i)-ypred(i)) 2 ; where ytrue(i) is the output data in the i-th control signal; ypred(i) is the predicted output data generated by the polynomial regression model from the input data in the i-th control signal; βmini is the sum of the squares of the differences between the output data and the predicted output data of N control signals, and N is the number of control signals.

[0078] Assume that the system input is voltage u(t) and the output is motor speed y(t). Select the polynomial regression model: y(t)=β0 + β1u(t)+β2u(t) 2, fitting the parameters by the gradient descent method, we get: β0 = 100, β1 = 2.5, β2 = -0.1.

[0079] Select the current operating point of the system (such as the equilibrium point u0) or the real-time input value u(t) as the expansion point. For the non-linear term (such as u(t) 2 ), perform Taylor expansion and retain up to the first-order term: u(t) 2 ≈u0 2 +2u0(u(t) - u0). Example: Expand u(t) in the polynomial model 2 at u0 = 5: u(t) 2 ≈5 2 +2×5×(u(t) - 5)=25 + 10(u(t) - 5), retain the first-order term in the Taylor expansion and ignore the higher-order terms (such as (u(t) - u0) 2 and higher-order terms).

[0080] Substitute the retained linear term into the original model to obtain the linearized expression. Example: Substitute the expansion into the polynomial model: y(t)≈β0 + β1u(t)+β2[25 + 10(u(t) - 5)]; after simplification, we get the linear function: ylin(t)=(β0 + 25β2)+(β1 + 10β2)u(t); substitute the parameter values: ylin(t)=(100 - 2.5)+(2.5 - 1)u(t)=97.5 + 1.5u(t), and obtain the current input value u(t) from the sensor or control system.

[0081] Substitute the input data u(t) in the last control signal into the linear function ylin(t) to calculate the predicted output or control quantity. Feed the intervention data back to the system and adjust the input to optimize the performance (such as reducing the error and improving the response speed). Example: Assume the input data u(t)=6 in the last control signal, substitute it into the linear function: ylin(t)=97.5 + 1.5×6 = 106.5. If the actual output ytrue(t)=105, then the error is 1.5, and the controller can adjust the input u(t) to reduce the error.

[0082] Specifically, perform PID control processing on at least one of the control signals to obtain the first intervention data, including: Construct a non-linear model based on at least one of the control signals; Call the non-linear model to generate the last expected output data according to the last input data; Obtain the output error signal according to the difference between the last output data and the last expected output data; Perform integral and differential operations on the output error signal to obtain an integral term and a differential term respectively; Generate first intervention data based on the weighted sum of a ratio, the integral term, and the differential term.

[0083] Exemplarily, perform polynomial regression on at least one of the control signals to construct a polynomial regression model, for example: y(t)=β0+β1u(t)+β2u(t) 2 +⋯+βnu(t) n ; where y(t) is the output data in the control signal; u(t) is the input data in the control signal, and β0, β1, β2, ⋯βn are the coefficients of each term in the initial nonlinear model respectively; n is any natural number greater than 2; Use methods such as the least squares method, the gradient descent method, or Bayesian optimization to minimize the prediction error: βmini=∑(ytrue(i)-ypred(i)) 2 ; where ytrue(i) is the output data in the i-th control signal; ypred(i) is the predicted output data generated by the polynomial regression model from the input data in the i-th control signal; βmini is the sum of the squares of the differences between the output data and the predicted output data of N control signals, and N is the number of control signals.

[0084] Assume that the system input is voltage u(t) and the output is motor speed y(t), and select the polynomial regression model: y(t)=β0+β1×u(t)+β2×u(t) 2 , and fit the parameters by the gradient descent method to obtain: β0 = 100, β1 = 2.5, β2 = -0.1.

[0085] Apply the constructed nonlinear model to the current input data (the last input data) to generate expected output data. Example: Assume the current input u(t)=6, substitute it into the model to get the expected output: yexpected(t)=100+2.5×6+(-0.1)×6 2 =111.4.

[0086] Obtain the actual output data (the last output data yactual(t)) from the system, calculate the difference between the actual output data and the expected output data, and obtain the output error signal: e(t)=yactual(t)-yexpected(t); Example: Assume the actual output yactual(t)=110, then the error signal is: e(t)=110 - 111.4 = -1.4.

[0087] Integrate the output error signal to obtain the integral term. The integration operation can be implemented by an accumulator, and the error value at each sampling point is multiplied by the sampling interval and then accumulated to obtain ∫e(t)dt; where dt is the sampling interval.

[0088] Differentiate the output error signal to obtain the differential term. The differentiation operation can be implemented by a subtractor and a divider, calculating the difference between adjacent sampling points and then dividing by the sampling interval: de(t) / dt ≈ [e(t) - e(t - Δt)] / Δt. Example: Assume the sampling interval Δt = 0.1 s, and perform integration and differentiation on the error signal: Integral term: ∫e(t)dt ≈ e(t)×Δt = -1.4×0.1 = -0.14; Differential term: de(t) / dt ≈ [(-1.4) - (-1.4)] / 0.1 = 0 (assuming the error at the previous sampling point is -1.4).

[0089] Generate intervention data based on the weighted sum of the proportional, integral, and differential terms. The formula for the weighted sum is: u(t) = Kpe(t) + Ki∫e(t)dt + Kdde(t) / dt; Apply the generated intervention data to the system to achieve real-time control or optimization. Example: Assume the proportional coefficient Kp = 1, the integral coefficient Ki = 0.5, and the differential coefficient Kd = 0, then the first intervention data is: u(t) = 1×(-1.4) + 0.5×(-0.14) + 0×0 = -1.47.

[0090] Specifically, perform state feedback control processing on at least one of the control signals to obtain the first intervention data, including: Construct a non-linear model based on at least one of the control signals; Identify at least one key state variable in the non-linear model; wherein, the key state variable is a state variable whose influence on the non-linear model exceeds a preset influence degree; Create a feedback gain matrix for the environmental control device according to the dynamic characteristic information and performance requirement information of the environmental control device; Obtain a feedback signal based on at least one of the key state variables and the feedback gain matrix; Superimpose the feedback signal and the last output data to generate a first intervention signal.

[0091] Exemplarily, perform polynomial regression on at least one of the control signals to construct a polynomial regression model, for example: y(t) = β0 + β1u(t) + β2u(t) 2 +⋯+βnu(t) n ; wherein, y(t) is the output data in the control signal; u(t) is the input data in the control signal, and β0, β1, β2, ⋯βn are the coefficients of each term in the initial non-linear model respectively; n is any natural number greater than 2; Use methods such as the least squares method, gradient descent method, or Bayesian optimization to minimize the prediction error: βmini = ∑(ytrue(i) - ypred(i))2 ; where, ytrue(i) is the output data in the i-th control signal; ypred(i) is the predicted output data generated by the input data in the i-th control signal through the polynomial regression model; βmini is the sum of the squares of the differences between the output data and the predicted output data of N control signals, and N is the number of control signals.

[0092] Assume that the system input is voltage u(t) and the output is motor speed y(t). Select the polynomial regression model: y(t) = β0 + β1u(t) + β2u(t) 2 , and fit the parameters by the gradient descent method to obtain: β0 = 100, β1 = 2.5, β2 = -0.1.

[0093] Use methods such as Sobol index and Morris method to quantify the influence degree of each state variable on the model output. Sobol index calculation formula: Si = Var(Y) / VarXi(EX∼i(Y∣Xi)); where, EX∼i(Y∣Xi) represents the conditional expectation of all other input variables X∼i (i.e., all variables except Xi) on the output variable Y under the condition that the i-th input variable Xi is fixed. That is to say, when Xi is fixed, the expected value of Y is only determined by other variables X∼i. This value reflects the average influence of other variables on Y when Xi remains unchanged.

[0094] VarXi(EX∼i(Y∣Xi)) represents the variance of the above conditional expectation EX∼i(Y∣Xi) with respect to Xi. That is to say, when Xi varies within its value range, the conditional expectation EX∼i(Y∣Xi) will also fluctuate. This variance quantifies the influence degree of the change of Xi on the conditional expectation.

[0095] Var(Y) represents the total variance of the output variable Y, that is, the uncertainty of Y. That is to say, this is the fluctuation degree of the model output Y under the combined action of all input variables.

[0096] Si represents the global sensitivity index of the i-th input variable Xi to the output Y, and its value is equal to the ratio of the numerator variance to the denominator total variance. The value range of Si is: 0 ≤ Si ≤ 1. Si → 0: Xi has almost no influence on Y. Si → 1: Almost all changes in Y are completely determined by Xi.

[0097] Select the state variables whose influence degree exceeds the preset threshold (such as 0.5) as key state variables.

[0098] Example: Assume the system state variables are x1(t) = u(t) and x2(t) = y(t). Using the Sobol index method, it is analyzed that: S1 = 0.8 (the influence degree of x1(t) on y(t)); S2 = 0.2 (the influence degree of x2(t) on y(t)). Setting the threshold to 0.5, then x1(t) is the key state variable.

[0099] Obtain the dynamic characteristic information (such as transfer function, state - space model, etc.) and performance requirement information (such as stability, response speed, etc.) of the environmental control device. Through the design of the feedback gain matrix, configure the poles of the closed - loop system at the desired positions to meet the performance requirements of the system. Use methods such as pole - placement algorithm, LQR (linear quadratic regulator) to calculate the feedback gain matrix. Example: Assume the system dynamic characteristic information is the transfer function G(s)=s + 11, and the performance requirement information is good stability and fast response speed. Using the pole - placement algorithm, configure the poles of the closed - loop system at s = - 2, and calculate the feedback gain matrix K = [1].

[0100] Use the key state variable and the feedback gain matrix to calculate the feedback signal. The calculation formula for the feedback signal is: ufb(t)= - K×x(t). Assume the current key state variable x1(t)=6, then the feedback signal is: ufb(t)= - K×x1(t)= - 1×6 = - 6.

[0101] Superimpose the feedback signal on the last - stage output data to generate the intervention signal. The calculation formula for the intervention signal is: u(t)=uff(t)+ufb(t). Example: Assume the feed - forward control signal uff(t)=5, then the intervention signal is: u(t)=uff(t)+ufb(t)=5+( - 6)= - 1.

[0102] In a preferred embodiment, perform strong non - linear processing on at least one of the device signals to obtain second intervention data, including: If it is determined that the environmental control device is of the fourth type, perform adaptive control processing on at least one of the device signals to obtain second intervention data; wherein, the fourth type indicates that the device signals generated by the environmental control device can establish a non - linear mathematical model; If it is determined that the environmental control device is of the fifth type, perform fuzzy control processing on at least one of the device signals to obtain second intervention data; wherein, the fifth type indicates that the device signals generated by the environmental control device cannot establish a mathematical model.

[0103] In this example, the adaptive control processing is a processing scheme that generates second intervention data by constructing a controlled system model and a reference model, and adjusting the controlled system model based on the parameter error between the reference model and the controlled system model.

[0104] Therefore, the adaptive controller can adjust parameters online to compensate for system uncertainties or time-varying characteristics, thereby improving the system's stability. By automatically adjusting the controller's parameters or structure, the adaptive controller can handle the uncertainties of the system model and external disturbances, enhancing the system's robustness. The adaptive controller can optimize the control strategy based on the current state of the system and the adjustment target to improve the system's performance.

[0105] The fuzzy control process is a processing solution that adjusts one or more output units in the output data based on at least one target input unit in the input data according to a preset fuzzy rule base to obtain second intervention data. Fuzzy control can handle uncertainties in the system, such as noise, interference, and model imprecision, improving the system's robustness and enabling it to adapt to different working environments and task requirements. Fuzzy control can utilize fuzzy rules to represent and process non-linear relationships in the system, achieving effective control of complex non-linear systems. It can adjust the fuzzy rule base according to different working environments and task requirements to improve the system's flexibility. Fuzzy control is based on a preset fuzzy rule base and does not require the establishment of an accurate mathematical model, simplifying the controller design process and reducing the controller's design cost and implementation complexity.

[0106] Specifically, performing adaptive control processing on at least one of the control signals to obtain second intervention data includes: Creating a controlled system model of the environmental control device based on at least one of the control signals; Creating a reference model of the environmental control device based on the historical information of the environmental control device; Invoking the reference model to generate expected output data based on the last input data; Generating a parameter error based on the expected output data and the last output data; the parameter error is used to characterize the deviation between the controlled system model and the reference model; Adjusting the controlled system model according to the parameter error so that the controlled system model generates second intervention data.

[0107] Exemplarily, creating the controlled system model of the environmental control device based on at least one of the control signals through a mechanism modeling method; wherein, it includes: by analyzing the physical structure and energy transfer relationship of the environmental control device (such as thermodynamic equations, fluid mechanics formulas), establishing an analytical relationship between the input (such as the control signal) and the output (such as temperature, humidity). For example: for the temperature control of an air conditioning system, the following model can be established: dt / dT = (Qin - Qout) / C; where, T is the temperature, Qin is the heating amount, Qout is the heat dissipation amount, and C is the heat capacity.

[0108] Create a controlled system model of the environmental control device according to at least one of the control signals through a data-driven modeling method; wherein, it includes: adopting an ARMA model (Autoregressive Moving Average Model) to describe the linear relationship between the output and historical inputs and outputs, and optimizing the model parameters by using the least squares method, gradient descent method or Bayesian optimization with measured input-output data. The Autoregressive Moving Average Model (ARMA model for short) is a classic statistical model in time series analysis. It combines the characteristics of the Autoregressive (AR) model and the Moving Average (MA) model and is used to describe and predict time series data.

[0109] Adopt ARIMA (Autoregressive Integrated Moving Average Model) or LSTM neural network to create a time series prediction model according to the historical information of the environmental control device as the reference model of the environmental control device.

[0110] Create a regression model according to the historical information of the environmental control device as the reference model of the environmental control device; wherein, it includes: establishing a multivariate linear regression or random forest model, analyzing the relationship between historical inputs (such as control signals, environmental parameters) and outputs (such as system states), and obtaining the reference model.

[0111] Input the last input data (such as the latest control signal and environmental parameters) into the reference model to generate the expected output data.

[0112] The parameter error includes one or more of the absolute error, relative error and root mean square error.

[0113] Absolute error = |last output data - expected output data|.

[0114] Relative error = (absolute error / |expected output data|) × 100%.

[0115] The root mean square error is the square root of the ratio of the square of the difference between the expected output data and the last output data to the number of times the last output data is collected, and is used to quantify the overall level of the prediction error.

[0116] Adjust the controlled system model according to the parameter error so that the controlled system model generates second intervention data, including: Based on the gradient descent method, perform iterative processing on the parameters of the controlled system model through a preset calculation loss function (such as the sum of squared parameter errors) to iteratively update the parameters of the controlled system model, so that the second prediction data of the controlled system model can reduce the parameter error; or Adopt the least mean square algorithm to adjust the coefficients of the controlled system model to minimize the prediction error: The formula for the gradient descent method is: θnew = θold - α∇θJ(θ); where θnew is the value of the parameter of the controlled system model before iterative update; θold is the value of the parameter of the controlled system model after iterative update; α is the learning rate; and ∇θJ(θ) is the calculation of the loss function.

[0117] The formula for the least mean square algorithm is: w(k + 1) = w(k) + μe(k)x(k); e(k) is the parameter error; w(k) is the value of the coefficient of the controlled system model before adjustment; w(k + 1) is the value of the coefficient of the controlled system model after adjustment; μ is the step size; and e(k) is the parameter error.

[0118] Specifically, performing fuzzy control processing on at least one of the device signals to obtain second intervention data includes: Determining a target input unit and a target output unit in at least one of the device signals through a preset fuzzy rule base; where the target input unit is one or more input units in the input data of the device signal; and the target output unit is one or more output units in the output data of the device signal; Determining an intervention coefficient of the environmental control device through the fuzzy rule base according to the target input unit and the target output unit; Invoking a preset intervention model to generate second intervention data according to the target output unit and the intervention coefficient.

[0119] Exemplarily, the target output unit is the control signal strength; the target input unit is the room temperature deviation; and the temperature deviation represents the range to which the difference between the current ambient temperature and the target temperature belongs; Example of the rule base: Rule number Control signal strength Temperature deviation Intervention coefficient R1 Low Very low deviation range 0.2 R2 Medium Lower deviation range 0.5 R3 High Normal deviation range 0.8 R4 Low Higher deviation range 1.2 R5 Medium Extremely high deviation range 1.5 The formula for the intervention model: I = K × S; where I: the second intervention data, that is, the adjusted strength of the target output unit, namely: the adjusted control signal strength; K is the intervention coefficient output by the fuzzy rule base; and S is the original target output unit, namely: the original control signal strength before adjustment.

[0120] In a preferred embodiment, sending the first intervention data or the second intervention data to the environmental control device as the output data of the environmental control device includes: Generating intervention indication information according to the first intervention data or the second intervention data; wherein, the intervention indication information is used to instruct the environmental control device to use the first intervention data or the second intervention data as the output data; Sending the intervention indication information to the environmental control device.

[0121] In this example, there is a timing asynchrony between manual intervention and system autonomous decision-making (such as the conflict between the emergency shutdown instruction and the PLC control cycle); for this, a time-triggered architecture (TTA) is adopted to encapsulate the intervention indication information as timestamped indication information to achieve nanosecond-level synchronization accuracy (compared with the traditional event-triggered architecture error > 10 ms).

[0122] Devices from different manufacturers use private protocols (such as Modbus / TCP vs EtherCAT), resulting in instruction set conflicts; for this, a semantic intermediate layer (SIL) is constructed to convert the intervention indication information into the IEC61131-3 standard IL instruction set to achieve cross-platform compatibility (supporting more than 20 mainstream industrial control protocols) Electromagnetic interference causes instruction packet loss (such as PWM signal interference in motor control); for this, a dual-redundancy transmission (DRT) technology is adopted to achieve dual transmission and reception of the intervention indication information at the Ethernet layer, reducing the bit error rate from 1e-5 to 1e-9 (meeting the IEC61508 SIL3 standard) Specifically, generating intervention indication information according to the first intervention data or the second intervention data includes: Verifying the first intervention data or the second intervention data to obtain a verification result; If the verification result is verification passed, performing data conversion on the first intervention data or the second intervention data to obtain conversion data; Generating intervention indication information according to the conversion data.

[0123] Exemplarily, verifying the first intervention data or the second intervention data through a configurable rule engine or a predefined verification template to obtain a verification result includes: Performing format verification on the first intervention data or the second intervention data to obtain a format verification result; for example: checking the field type (such as string, numeric, date), length limit (such as 11 digits for a mobile phone number), and regular expression matching (such as email format).

[0124] Perform business logic verification on the first intervention data or the second intervention data to obtain a logic verification result; for example: verify data relevance (such as the order amount not exceeding the inventory value), status consistency (such as the operation permission matching the user role).

[0125] Perform range constraint verification on the first intervention data or the second intervention data to obtain a range verification result; for example: threshold check of numerical fields (such as the temperature value being between -20°C and 50°C).

[0126] If it is determined that the format verification result, the logic verification result, and the range verification result are all verified to pass, then generate a verification result with the content of verification passed.

[0127] Perform data conversion on the first intervention data or the second intervention data to obtain conversion data, including: If it is determined that there is a long text field in the first intervention data or the second intervention data, then compress the long text field to obtain a short text field; for example: perform lossless compression on the long text field (such as using the LZ4 algorithm) to reduce the transmission volume.

[0128] Perform structured processing on the first intervention data or the second intervention data to obtain structured data, and use the structured data as the conversion data; for example: generate structured data (such as JSON / XML) or binary stream for subsequent steps.

[0129] Generate intervention indication information based on the conversion data, including: Define a frame start identifier (such as the magic number 0xAA 0x55).

[0130] Add a protocol version (such as v1.2), a message type (such as an intervention instruction), and a data length field.

[0131] Insert a timestamp (Unix timestamp or high-precision time) to obtain a frame header.

[0132] Supplement context information to obtain metadata; the metadata includes one or several of: data source identifier (such as the device ID of an environmental control device), operator information, priority marker (such as urgent / normal), data hash value (for anti-tampering).

[0133] Reserve a CRC32 placeholder (to be filled after subsequent calculation).

[0134] Use a predefined.proto file to describe the data structure (such as message Intervention {...}).

[0135] Serialize the converted data into a binary stream according to the rules to obtain serialized data, which supports efficient compression and cross-platform parsing.

[0136] Calculate the CRC32 check value for the binary stream after concatenating the frame header, metadata, and serialized data, and replace the reserved placeholder in the metadata with the calculation result.

[0137] After concatenating the frame header, the metadata, the serialized data, and the CRC32 check value, add a frame end identifier (such as 0x55 0xAA) to obtain intervention indication information.

[0138] Embodiment 2: Please refer to Figure 2 , a control device 2 for an environmental control device in a processing workshop according to this embodiment includes: An input module 21, configured to receive at least one device signal sent by an environmental control device; wherein, the device signal includes an input data and an output data; the input data is used to characterize the environmental condition of the space where the environmental control device is located or the operating condition of the environmental control device at a time point; the output data is control data for the environmental control device to change or maintain the environmental condition or the operating condition at a time point. A category recognition module 22, configured to construct a non-linear model according to at least one of the device signals, and determine the system category of the environmental control device according to the non-linear model; wherein, the non-linear model reflects the non-linear degree of the environmental control device; the system category includes weak non-linearity and strong non-linearity. A weak non-linearity module 23, configured to perform weak non-linearity processing on at least one of the device signals if the category of the environmental control device is a weak non-linearity category to obtain first intervention data; the weak non-linearity processing is one of linearization processing, PID control processing, and state feedback control processing; the first intervention data is used as the output data of the environmental control device to make the input data reach a preset target index. A strong non-linearity module 24, configured to perform strong non-linearity processing on at least one of the device signals if the category of the environmental control device is a strong non-linearity category to obtain second intervention data; wherein, the strong non-linearity processing is adaptive control processing or fuzzy control processing; the second intervention data is used as the output data of the environmental control device to make the input data reach a preset target index. An output module 25, configured to send the first intervention data or the second intervention data to the environmental control device.

[0139] Embodiment 3: To achieve the above object, the present invention further provides an electronic device 3. The components of the control method device of the environmental control device in Embodiment 3 can be dispersed in different electronic devices. The electronic device 3 can be a smart phone, a tablet computer, a notebook computer, a desktop computer, a rack server, a blade server, a tower server or a cabinet server (including an independent server, or a server cluster composed of multiple application servers) that executes a program, etc. The electronic device in this embodiment at least includes, but is not limited to: a memory 31 and a processor 32 that can be communicatively connected to each other through a system bus, as Figure 3 shown. It should be noted that Figure 3 only an electronic device with components - is shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented.

[0140] In this embodiment, the memory 31 (i.e., the readable storage medium) includes flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 31 can be an internal storage unit of the electronic device, such as the hard disk or memory of the electronic device. In other embodiments, the memory 31 can also be an external storage device of the electronic device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc., equipped on the electronic device. Of course, the memory 31 can also include both the internal storage unit and the external storage device of the electronic device. In this embodiment, the memory 31 is generally used to store the operating system and various application software installed in the electronic device, such as the program code of the control method device of the environmental control device in Embodiment 3. In addition, the memory 31 can also be used to temporarily store various data that have been output or will be output.

[0141] The processor 32 can be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 32 is generally used to control the overall operation of the electronic device. In this embodiment, the processor 32 is used to run the program code stored in the memory 31 or process data, such as running the control method device of the environmental control device to implement the control methods of the environmental control device in Embodiment 1 and Embodiment 2.

[0142] Embodiment 4: To achieve the above object, the present invention further provides a computer-readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, server, App application store, etc., on which a computer program is stored, and when the program is executed by the processor 32, the corresponding functions are implemented. The computer-readable storage medium of this embodiment is used to store a computer program for implementing the control method of the environmental control device, and when executed by the processor 32, it implements the control methods of the environmental control device in Embodiment 1 and Embodiment 2.

[0143] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0144] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases, the former is a better implementation method.

[0145] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A control method for an environmental control device in a processing workshop, characterized in that, Including: Receiving at least one device signal sent by an environmental control device; wherein, the device signal includes an input data and an output data; Constructing a non-linear model based on at least one of the device signals, and determining the system category of the environmental control device according to the non-linear model; wherein, the non-linear model reflects the non-linear degree of the environmental control device; the system category includes weak non-linearity and strong non-linearity; If the category of the environmental control device is a weak non-linear category, performing weak non-linear processing on at least one of the device signals to obtain first intervention data; wherein, the first intervention data is used as the output data of the environmental control device to make the input data reach a preset target index; If the category of the environmental control device is a strong non-linear category, performing strong non-linear processing on at least one of the device signals to obtain second intervention data; wherein, the second intervention data is used as the output data of the environmental control device to make the input data reach a preset target index; Sending the first intervention data or the second intervention data to the environmental control device.

2. The control method according to claim 1, characterized in that The input data is used to characterize the environmental condition of the space where the environmental control device is located or the operating condition of the environmental control device at a time point; the output data is the control data for the environmental control device to change or maintain the environmental condition or the operating condition at a time point; The weak non-linear processing is one of linearization processing, PID control processing, and state feedback control processing; The strong non-linear processing is adaptive control processing or fuzzy control processing.

3. The control method according to claim 1, wherein Receiving at least one device signal sent by an environmental control device includes: Sending an interaction instruction to an environmental control device; wherein, the interaction instruction has any one of a time signal, a time period signal, and a quantity signal; the time signal has a time value, and the time signal is used to instruct the environmental control device to send the device signal within the time value; the time period signal has at least one time period, and the time period signal is used to instruct the environmental control device to send the device signal within the time period; the quantity signal has a quantity value, and the quantity signal is used to instruct the environmental control device to send the device signal of the quantity value; Receiving at least one device signal sent by the environmental control device according to the interaction instruction.

4. The control method according to claim 1, characterized in that, Constructing a non-linear model based on at least one of the device signals includes: Constructing an analysis model of the environmental control device based on at least one of the device signals; Determining the lag order in the analysis model, and selecting a basis function according to system characteristic information; Based on the analysis model, constructing the non-linear model of the environmental control device according to the lag order and the basis function.

5. The control method according to claim 1, characterized in that Determining the system category of the environmental control device according to the non-linear model includes: Combining at least one basis function in the non-linear model into a regression matrix; Converting the non-linear model into a linear function according to the regression matrix; Determine the non - linear contribution rate of the environmental control device according to the linear function; wherein, the non - linear contribution rate characterizes the importance of the non - linear term in the non - linear model in the non - linear model Determine the system category of the environmental control device according to the non - linear contribution rate.

6. The control method according to claim 2, wherein Perform weak non - linear processing on at least one of the device signals to obtain first intervention data, including:[[]] If it is determined that the environmental control device is of the first type, perform linearization processing on at least one of the device signals to obtain first intervention data; wherein, the first type indicates that the output data of the environmental control device has one output variable, and by adjusting the output data of the environmental control device within a first range near a working point, the input data of the environmental control device reaches a preset first input index; If it is determined that the environmental control device is of the second type, perform PID control processing on at least one of the device signals to obtain first intervention data; wherein, the second type indicates that the output data of the environmental control device has one output variable, and it is necessary to adjust the output data of the environmental control device within a preset second time so that the input data of the environmental control device reaches a preset second input index; If it is determined that the environmental control is of the third type, perform state - feedback control processing on at least one of the device signals to obtain first intervention data; wherein, the third type indicates that the output data of the environmental control device has two or more output variables, and it is necessary to adjust at least one of the output variables so that the input data of the environmental control device reaches a preset third input index.

7. The control method according to claim 2, wherein Perform strong non - linear processing on at least one of the device signals to obtain second intervention data, including:[[]] If it is determined that the environmental control device is of the fourth type, perform adaptive control processing on at least one of the device signals to obtain second intervention data; wherein, the fourth type indicates that the device signal generated by the environmental control device can be linearly processed; If it is determined that the environmental control device is of the fifth type, perform fuzzy control processing on at least one of the device signals to obtain second intervention data; wherein, the fifth type indicates that the device signal generated by the environmental control device cannot establish a mathematical model.

8. A control device for an environmental control device in a processing workshop, characterized in that, Including:[[]] An input module for receiving at least one device signal sent by an environmental control device; wherein, the device signal includes an input data and an output data; A category recognition module for constructing a non - linear model according to at least one of the device signals and determining the system category of the environmental control device according to the non - linear model; wherein, the non - linear model reflects the non - linear degree of the environmental control device; the system category includes weak non - linearity and strong non - linearity; A weak non - linear module, which is used to perform weak non - linear processing on at least one of the device signals if the category of the environmental control device is a weak non - linear category, and obtain first intervention data; wherein, the weak non - linear processing is one of linearization processing, PID control processing, and state feedback control processing; the first intervention data is used as the output data of the environmental control device to make the input data reach a preset target index; A strong non - linear module, which is used to perform strong non - linear processing on at least one of the device signals if the category of the environmental control device is a strong non - linear category, and obtain second intervention data; wherein, the strong non - linear processing is adaptive control processing or fuzzy control processing; the second intervention data is used as the output data of the environmental control device to make the input data reach a preset target index; An output module, which is used to send the first intervention data or the second intervention data to the environmental control device.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor of the electronic device executes the computer program, it implements the steps of the control method of the environmental control device according to any one of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, characterized in that, When the computer program stored in the readable storage medium is executed by the processor, it implements the steps of the control method of the environmental control device according to any one of claims 1 to 7.

Images