State Prediction Method and Device Based on Restoration of Physical System State

Abstract

The present application discloses a state prediction method and device based on physical system state restoration. By applying the technical solution of the present application, in the sparse identification algorithm of the nonlinear dynamic system used in restoring the physical system in a certain data recognition scenario, a method of adaptively selecting a threshold for the nonlinear dynamic system can be achieved by adding a method of constructing a correlation function B with the threshold λ as an independent variable, so that the subsequent nonlinear dynamic system configured with a self-selected threshold is used to restore the state information of a physical system within a period of time. Thus, an object of correctly restoring the physical system state without manual threshold selection is achieved. It also solves the reduction error influence caused by missing function terms that may still occur in the process of using the nonlinear dynamic system to restore the physical system in the related art even when a relatively small threshold is selected.

Description

Technical Field

[0001] The present application relates to analysis data processing technology, in particular to a state prediction method and device based on physical system state restoration. Background Art

[0002] In recent years, with the rapid development of science and technology, the application research of nonlinear dynamic systems has received more and more extensive attention in various fields. As a powerful supplement to traditional linear system models, the research on nonlinear dynamic system models will strongly promote scientific research and industrial development in various fields.

[0003] In related technologies, the sparse identification method of nonlinear dynamic systems is often used to restore the states of some physical systems in data identification scenarios. For example, this sparse identification method can be used to identify the state of a meteorological system in a meteorological data scenario. Or this sparse identification method can also be used to identify the state of a traffic system in a traffic data scenario, etc. Among them, in the sparse identification algorithm of nonlinear dynamic systems, a threshold λ needs to be artificially selected, and the role of this threshold is to ensure that the solution of the nonlinear dynamic system is sparse. Further, the way to artificially select this threshold is to select a threshold smaller than the minimum absolute value of the coefficients for restoring the physical system. However, if the coefficients of the restored dynamic system are not understood, in more cases, a relatively small threshold will be directly selected.

[0004] However, in related technologies, when using the sparse identification method of nonlinear dynamic systems to restore physical systems, even when a relatively small threshold λ is artificially selected, the problem that the restoration result of the physical system still has a large error cannot be avoided. Therefore, how to implement a technical solution for automatically selecting a suitable threshold for a nonlinear dynamic system to correctly restore the state of the physical system has become a problem that needs to be solved by those skilled in the art. Summary of the Invention

[0005] The embodiments of the present application provide a state prediction method and device based on physical system state restoration, and the embodiments of the present application are used to solve the problem in related technologies that a suitable threshold cannot be selected for a nonlinear dynamic system, resulting in the inability to correctly restore the state of the physical system.

[0006] Among them, according to one aspect of the embodiments of the present application, a state prediction method based on physical system state restoration is provided, and the method includes:

[0007] Determine a physical system to be restored existing in the real environment, and create a nonlinear dynamic system for capturing the nonlinear characteristics of the physical system to be restored, where the nonlinear characteristics are used to reflect the real behavior of the physical system to be restored;

[0008] Calculate a first function and a second function for the non - linear dynamic system, where the first function is used to reflect the functional relationship of the data error between the simulated data and the real data of the restored physical system, and the second function is used to reflect the functional relationship between the threshold and the sparsity of the restored physical system. The restored physical system is the physical system whose state needs to be restored, and the threshold

[0009] is used to control the sparsity degree of the output value of the non - linear dynamic system; value;

[0010] Use the first function and the first function to calculate a correlation function for the non - linear dynamic system, and use the correlation function to select a corresponding target value for the non - linear dynamic system;

[0011] Configure the target

[0012] value as the self - selection threshold of the non - linear dynamic system, and use the non - linear dynamic system configured with the self - selection threshold to perform state restoration on the restored physical system in the data recognition scenario for a preset duration to obtain the target physical system after state restoration;

[0013] L ;

[0014] where the L is the first function, which is used to reflect the functional relationship between the real data and the fitted data.

[0015] Optionally, in another embodiment based on the above - mentioned method of the present application, the first function is calculated for the non - linear dynamic system through the following formula:

[0016] ;

[0017] where, the is the second function, the is the number of all non - zero terms in the function library of the restored physical system, and the is the number of all column terms in the function library.

[0018] Optionally, in another embodiment of the method based on the present application, the correlation function is calculated for the non-linear dynamic system by the following formula:

[0019] ;

[0020] wherein the is the correlation function, the L is the first function, and the is the second function.

[0021] Optionally, in another embodiment of the method based on the present application, the step of selecting a corresponding target value for the non-linear dynamic system by using the correlation function includes:

[0022] Taking the value corresponding to the maximum value range of the correlation function value in the correlation function as the target value.

[0023] Optionally, in another embodiment of the method based on the present application, the step of configuring the target value as the self-selection threshold of the non-linear dynamic system includes:

[0024] Taking the target value as the self-selection threshold of the sparse identification algorithm of the non-linear dynamic system;

[0025] wherein the sparse identification algorithm is obtained by the following formula:

[0026] ;

[0027] wherein, is the sparse coefficient, the is the residual term of the sparse identification algorithm, the is the regularization term of the sparse identification algorithm, and y is the real data.

[0028] Optionally, in another embodiment of the method based on the present application, the step of using the non-linear dynamic system configured with the self-selection threshold to perform state restoration on the restored physical system in a data identification scenario for a preset duration includes:

[0029] Using the non-linear dynamic system configured with the self-selection threshold to perform state restoration on the test physical system to obtain a test restoration result; and collecting the test real result corresponding to the test physical system in the real situation;

[0030] When the error value between the detected test reduction result and the test true result is less than a preset difference value, the physical system to be restored in the data recognition scenario is restored in state within a preset time period by using the non-linear dynamic system configured with the self-selection threshold.

[0031] Wherein, according to another aspect of the embodiments of the present application, a state prediction device based on the state restoration of a physical system is provided, including:

[0032] A calculation module, configured to calculate a first function and a second function for the non-linear dynamic system, wherein the first function is a function relationship reflecting the data error between the simulated data and the real data of the physical system to be restored, and the second function is used to reflect the threshold The function relationship between the sparsity between the threshold and the physical system to be restored, and the physical system to be restored is the physical system that needs to be restored in state, and the threshold Is used to control the sparsity degree of the output value of the non-linear dynamic system;

[0033] A selection module, configured to calculate a correlation function for the non-linear dynamic system by using the first function and the first function, and select a corresponding target for the non-linear dynamic system by using the correlation function Value;

[0034] An analysis module, configured to configure the target Value as the self-selection threshold of the non-linear dynamic system, and use the non-linear dynamic system configured with the self-selection threshold to restore the state of the physical system to be restored in the data recognition scenario within a preset time period.

[0035] According to another aspect of the embodiments of the present application, an electronic device is provided, including:

[0036] A memory, configured to store executable instructions; and

[0037] A display, configured to execute the executable instructions with the memory to complete the operations of any of the above state prediction methods based on the state restoration of the physical system.

[0038] According to still another aspect of the embodiments of the present application, a computer-readable storage medium is provided, configured to store instructions readable by a computing device, and when the instructions are executed, the operations of any of the above state prediction methods based on the state restoration of the physical system are executed.

[0039] In the present application, a first function and a second function are calculated for the non-linear dynamic system, wherein the first function is a function relationship reflecting the data error between the simulated data and the real data of the physical system to be restored, and the second function is used to reflect the threshold The functional relationship between the sparsity and the physical system to be restored, where the physical system to be restored is the physical system for which state restoration is required, and the threshold used to control the sparsity degree of the output value of the non-linear dynamic system; by using the first function and the second function, a related function is calculated for the non-linear dynamic system, and the corresponding target value; the target value is configured as the self-selection threshold of the non-linear dynamic system, and the non-linear dynamic system with the configured self-selection threshold is used to perform state restoration on the physical system to be restored in the data recognition scenario for a preset duration.

[0040] By applying the technical solution of the present application, in the sparse recognition algorithm of the non-linear dynamic system used in the physical system to be restored in a certain data recognition scenario, a method of adaptively selecting a threshold for the non-linear dynamic system can be achieved by adding a method of constructing a related function B with the threshold λ as an independent variable, so that the non-linear dynamic system with the configured self-selection threshold is used to restore the state information of a certain physical system within a period of time. Thus, an object of correctly restoring the state of the physical system without manual threshold selection is achieved. It also solves the reduction error influence caused by missing function terms that may still occur in the process of using the non-linear dynamic system to restore the physical system in the related art even when a relatively small threshold is selected.

[0041] The following uses multiple embodiments to further describe the technical solution of the present application in detail. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings forming a part of the specification depict embodiments of the present application and, together with the description, are used to explain the principles of the present application.

[0043] Referring to the drawings, the present application can be more clearly understood according to the following detailed description, where:

[0044] Figure 1 shows a schematic diagram of the system architecture of a state prediction based on physical system state restoration provided by an embodiment of the present application;

[0045] Figure 2 shows a schematic diagram of the related function B provided by an embodiment of the present application;

[0046] Figure 3 shows a schematic diagram of the overall process of a state prediction based on physical system state restoration provided by an embodiment of the present application;

[0047] Figure 4 shows a schematic diagram of the structure of an electronic device provided by an embodiment of the present application;

[0048] Figure 5 shows a schematic structural diagram of an electronic device provided by an embodiment of the present application;

[0049] Figure 6 shows a schematic diagram of a storage medium provided by an embodiment of the present application. Detailed implementation manners

[0050] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that: unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present application.

[0051] Meanwhile, it should be understood that, for the sake of convenience of description, the sizes of the various parts shown in the drawings are not drawn in actual proportional relationships.

[0052] The following description of at least one exemplary embodiment is merely illustrative in nature and is not intended as any limitation to the present application and its application or use.

[0053] Technologies, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and devices should be regarded as part of the specification.

[0054] It should be noted that: like reference numerals and letters denote like items in the following drawings, and thus, once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0055] In addition, the technical solutions between various embodiments of the present application may be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present application.

[0056] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present application are only used to explain the relative positional relationship and movement conditions between components in a specific posture (as shown in the drawings). If the specific posture changes, the directional indications will also change accordingly.

[0057] The following is combined with Figures 1 - 3 to describe a state prediction method for performing physical system state restoration according to an exemplary embodiment of the present application. It should be noted that the following application scenarios are only shown for the convenience of understanding the spirit and principle of the present application, and the embodiments of the present application are not limited in this regard. On the contrary, the embodiments of the present application can be applied to any applicable scenario.

[0058] Furthermore, the present application also proposes a state prediction method and apparatus based on physical system state restoration.

[0059] Figure 1 Schematically shown is a flowchart of a state prediction method based on physical system state restoration according to an embodiment of the present application. As Figure 1 shown, the method includes:

[0060] S101, determining a restoration physical system to be restored existing in the real environment, and creating a non - linear dynamic system for capturing the non - linear characteristics of the restoration physical system, where the non - linear characteristics are used to reflect the real behavior of the restoration physical system;

[0061] Among them, the restoration physical system refers to a physical system that already exists in the real environment but needs to restore or simulate its state through a certain method or technology. This can be a specific mechanical system, ecological system, climate system, etc. Different from a linear system, the behavior of a non - linear dynamic system cannot be described by a linear equation. They may contain non - linear factors such as exponential growth, logarithmic relationship, polynomial relationship, etc. Non - linear dynamic systems are often used to simulate complex phenomena in the real world. Non - linear characteristics refer to the non - linear factors existing in the system behavior, which make the relationship between the output and input of the system not proportional. Non - linear characteristics can reflect the real behavior of the system, including its complexity and unpredictability.

[0062] In some embodiments of the present application, assume there is a traffic system in a city. This system can be regarded as a "restoration physical system" because it actually exists and has physical entities (such as vehicles, roads, etc.). To simulate this traffic system, a "non - linear dynamic system" capable of capturing non - linear characteristics (such as traffic flow, vehicle density, and traffic patterns, etc.) needs to be created.

[0063] S102, calculating a first function and a second function for the non - linear dynamic system, where the first function is used to reflect the functional relationship of the data error between the simulated data and the real data of the restoration physical system, and the second function is used to reflect the functional relationship between the threshold and the sparsity of the restoration physical system. The restoration physical system is the physical system that needs to perform state restoration, and the threshold is used to control the sparsity degree of the output value of the non - linear dynamic system.

[0064] Among them, the threshold value λ is a parameter used to control the sparsity of the model output. In the sparse recognition algorithm, the threshold determines which parameters will be considered important and retained in the model, and which will be considered unimportant and set to zero. The data error refers to the difference between the simulated data (the data obtained through model prediction) and the real data (the actually observed data). The data error is an important indicator for measuring the prediction accuracy of the model. Sparsity refers to a vector, matrix, or other data structure in which most elements are zero. In machine learning and signal processing, sparse models are usually more preferable because they are simpler and easier to interpret.

[0065] In the related technologies, a dynamic system is an important concept in mathematics used to describe the evolution of certain objects over time, usually described by differential equations.

[0066] Furthermore, a nonlinear dynamic system refers to a dynamic system that contains nonlinear function terms. The nonlinear function terms include some basic elementary functions such as exponential functions, power functions, logarithmic functions, polynomial functions, trigonometric functions, etc. and composite functions formed by their combinations.

[0067] Specifically, a nonlinear dynamic system refers to a dynamic system whose some properties do not have linear characteristics. Different from linear systems, the evolution law of nonlinear systems is not only affected by external factors, but also by internal factors of the system. Compared with the advantages of linear systems such as predictability and certainty, nonlinear dynamic systems have characteristics such as unpredictability and being far from the equilibrium state.

[0068] In one way, the modeling of nonlinear dynamic systems is the core content of the research on nonlinear dynamic systems. For different research objects, the modeling of nonlinear dynamic systems can be divided into multiple fields such as physical modeling, social modeling, and economic modeling. For physical modeling, the nonlinear dynamic system model can be used to describe different fields such as astrophysics, fluid mechanics, and material mechanics; for the modeling of social engineering, the nonlinear dynamic system model can be used to simulate social structure changes and social network trends, etc.

[0069] In the related technologies, when establishing a nonlinear dynamic system model, the most crucial thing is to determine appropriate control variables and state variables, seriously study the interactions between systems, and understand the relationships between parameters in the system in order to construct a reasonable and effective model. At the same time, this process also needs to consider the sensitivity of the system to model parameters and its corresponding changes in response to various initial conditions.

[0070] In addition, with the continuous development of the modeling technology of nonlinear dynamic systems, its applications in various fields are becoming more and more extensive.

[0071] In a certain scenario, business personnel can use a non-linear dynamic system to restore the operating state of a physical system in a data recognition scenario.

[0072] For example, the data recognition scenario can be a meteorological recognition scenario, a traffic recognition scenario, a financial recognition scenario, a human resource recognition scenario, etc.

[0073] Specifically, it can be to solve a given function value for the physical system to be measured through the sparse recognition algorithm of the non-linear dynamic system, and this given function value is used to reflect: the change of the state value of the physical system over a certain period of time.

[0074] In the related art, in the sparse recognition algorithm of the non-linear dynamic system, a threshold λ needs to be manually selected. The role of this threshold is to ensure that the solution of the sparse recognition algorithm is sparse. Further, the method for manually selecting this threshold is: first determine the minimum value of the absolute value of the coefficients of the restored physical system, and select a value smaller than this minimum value as the threshold λ.

[0075] It can be understood that if the business personnel do not understand the coefficients of the restored physical system, in more cases, a relatively small value will be directly selected as the threshold λ.

[0076] However, when using the sparse recognition method of the non-linear dynamic system to restore the physical system, in the case of selecting a relatively small threshold λ, it is still inevitable that there will be a large error in the restoration result of the physical system, because it is possible to ignore the function terms with smaller coefficients.

[0077] Based on this, the present application adopts a method of automatically selecting the threshold by using a correlation function (i.e., the B function) with the constructed threshold λ as the independent variable, so as to avoid the problem of possible errors in restoring the physical system when selecting a relatively small threshold λ. Furthermore, the purpose of more accurately restoring the state of the physical system is achieved.

[0078] Further, the embodiment of the present application needs to first prepare a data set:

[0079] As an example, the data set X is:

[0080] ;

[0081] All parameters in X are the values of all dependent variables of the physical system (i.e., the restored physical system) that needs to perform state restoration within a preset time period, so as to perform state restoration of the restored physical system within a preset time period based on this data set. are multiple time points, and X is composed of row vectors , , ,…, Composed, each row vector represents a time point of data, each row vector contains n elements , , , , and these elements are the time points of the values of n variables, where is the transpose operation, m is the number of time points, and n is the number of variables.

[0082] It should be noted that when the collected data set includes noisy data with obvious noise levels, it is also necessary to perform a certain degree of noise reduction on the data and take the first derivative of the data set with respect to time numerically.

[0083] Furthermore, the embodiments of the present application also need to construct a basic function library:

[0084] As an example, the basic function library is:

[0085] , and this basic function library can be composed of polynomials , trigonometric functions , exponential functions , logarithmic functions, etc. In one way, the embodiments of the present application can store all the basic functions in the basic function library and apply the previously collected data set to the constructed basic function library to obtain a specific data basic function library.

[0086] Further, the embodiments of the present application need to first construct a first function (i.e., the L function) and a second function (i.e., the S function).

[0087] Among them, the L function (i.e., the first function) is the log function relationship between the restored physical system and the error of the real data. And the S function (i.e., the second function) is used to reflect the relationship between the threshold λ and the sparsity of the restored physical system.

[0088] S103, using the first function and the second function, calculate the relevant function for the nonlinear dynamic system, and use the relevant function to select the corresponding target value for the nonlinear dynamic system.

[0089] In one way, the embodiments of the present application can combine the L function (the first function) and the S function (the second function) using a linear combination to form the relevant function B of the nonlinear dynamic system.

[0090] As an example, the formula definitions of the respective functions are as follows:

[0091] In the first aspect, the first function L ;

[0092] wherein is a function for reflecting the functional relationship between the real data and the fitted data. The calculation formula of the E function is:

[0093] .

[0094] In the second aspect, the second function ;

[0095] wherein, the is the number of all non-zero terms in the function library in the restored physical system, is the number of all column terms in the function library.

[0096] In the third aspect, the correlation function .

[0097] As an example, as Figure 2 shown, it is a schematic diagram of the correlation function B obtained in the embodiment of the present application. In one way, when the correlation function B in the embodiment of the present application reaches the maximum value, the minimum value in the interval value of the corresponding threshold λ can be used as the target value.

[0098] For example, as Figure 2 visible, when the correlation function B reaches the maximum value of 0, the interval value of the corresponding threshold λ is 0 - 0.6, and then the minimum value 0 in this interval can be used as the target value.

[0099] S104, configure the target value as the self-selection threshold of the non-linear dynamic system, and use the non-linear dynamic system with the configured self-selection threshold to restore the state of the restored physical system in the data recognition scenario for a preset duration, so as to obtain the target physical system after state restoration.

[0100] In one way, after determining the target value in the embodiment of the present application, it can be used as the self-selection threshold to restore the state of the physical system within the preset duration.

[0101] As an example, in the embodiment of the present application, the basic structure for restoring the meteorological physical system can be determined according to the following data:

[0102] ;

[0103] ;

[0104] A ;

[0105] ;

[0106] Among them, represents the first-order derivative of the variable with respect to time; represents the basic function library of the meteorological system; is the corresponding coefficient matrix.

[0107] Furthermore, in the non-linear dynamic system, the calculation formula for solving the sparse coefficient matrix in the sparse identification algorithm is as follows:

[0108] ;

[0109] Among them, is the sparse coefficient, is the residual term of the sparse identification algorithm, is the regularization term of the sparse identification algorithm, and y is the real data.

[0110] It can be understood that the threshold λ controls the sparsity of the solution.

[0111] Furthermore, after the embodiment of the present application determines the target value, the coefficient matrix of the meteorological physical system can be obtained by solving using the above formula. Furthermore, the state parameters of the meteorological physical system can be obtained accordingly.

[0112] By applying the technical solution of the present application, in the sparse identification algorithm of the non-linear dynamic system used in restoring the physical system in a certain data recognition scenario, a method of constructing the related function B using the threshold λ as the independent variable can be added to adaptively select the threshold for the non-linear dynamic system, so that the non-linear dynamic system configured with the self-selection threshold is used to restore the state information of a certain physical system within a period of time. Thus, an object of correctly restoring the state of the physical system without manual selection of the threshold is achieved. It also solves the reduction error influence caused by missing function terms that may still occur in the process of using the non-linear dynamic system to restore the physical system in the related technology even when a relatively small threshold is selected.

[0113] Optionally, in another embodiment based on the above method of the present application, the first function is calculated for the non-linear dynamic system through the following formula:

[0114] L ;

[0115] Among them, the L is the first function, which is used to reflect the functional relationship between the real data and the fitting data.

[0116] Optionally, in another embodiment based on the above method of the present application, the second function is calculated for the non-linear dynamic system through the following formula:

[0117] ;

[0118] wherein, the is the second function, the is the number of all non-zero terms in the function library of the reduced physical system, and the is the number of all column terms in the function library.

[0119] Optionally, in another embodiment based on the above method of the present application, the correlation function is calculated for the non-linear dynamic system through the following formula:

[0120] ;

[0121] wherein, the is the correlation function, the L is the first function, and the is the second function.

[0122] Optionally, in another embodiment based on the above method of the present application, the step of using the correlation function to select a corresponding target value for the non-linear dynamic system includes:

[0123] taking the value corresponding to the maximum value range of the correlation function value in the correlation function as the target value.

[0124] Optionally, in another embodiment based on the above method of the present application, the step of configuring the target value as the self-selection threshold of the non-linear dynamic system includes:

[0125] taking the target value as the self-selection threshold of the sparse identification algorithm of the non-linear dynamic system;

[0126] wherein, the sparse identification algorithm is obtained by the following formula:

[0127] ;

[0128] wherein, is the sparse coefficient, the is the residual term of the sparse identification algorithm, the is the regularization term of the sparse identification algorithm, and y is the true data.

[0129] Optionally, in another embodiment based on the above method of the present application, the state reduction of the restored physical system in the data recognition scenario for a preset duration by using the nonlinear dynamic system configured with the self-selection threshold includes:

[0130] Using the nonlinear dynamic system configured with the self-selection threshold to perform state reduction on the test physical system to obtain a test reduction result; and collecting the test true result corresponding to the test physical system under real conditions;

[0131] When it is detected that the error value between the test reduction result and the test true result is less than a preset difference, use the nonlinear dynamic system configured with the self-selection threshold to perform state reduction on the restored physical system in the data recognition scenario for a preset duration.

[0132] In one way, as Figure 3 shown, a schematic diagram of the overall state prediction process based on the state reduction of the physical system provided by an embodiment of the present application is shown, which includes:

[0133] Step 1: Collect a data set for restoring the physical system and numerically calculate the first derivative of the data set with respect to time.

[0134] Among them, the first derivative of the data set with respect to time is expressed as:

[0135] .

[0136] Step 2: Obtain a basic function library and use this data set to construct parameters for the basic function library to obtain a data basic function library.

[0137] Among them, the basic function library is expressed as:

[0138] A

[0139] Step 3: Calculate the first function L that reflects the functional relationship of the data error between the restored physical system and the real data.

[0140] Among them, the first function L is expressed as:

[0141] L ;

[0142] Among them, reflects the functional relationship between the real data and the fitted data.

[0143] Step 5: Calculate the second function S that reflects the functional relationship between the threshold and the sparsity of the restored physical system.

[0144] Among them, the second function S is expressed as:

[0145] ;

[0146] wherein, the is the number of all non - zero terms in the function library of the reduction physical system, and the is the number of all column terms in the function library.

[0147] Step 5: Use the difference between the first function L and the first function S to calculate the correlation function B for the non - linear dynamic system.

[0148] wherein, the correlation function B is expressed as:

[0149] ;

[0150] wherein, L is the first function, is the second function.

[0151] Step 6: Determine the value interval corresponding to the maximum value interval of the correlation function value in the correlation function, and take the minimum value in the value interval as the target value.

[0152] Step 7: Take the target value as the self - selection threshold of the sparse identification algorithm of the non - linear dynamic system, and use the non - linear dynamic system configured with the self - selection threshold to restore the state of the reduction physical system within a preset duration in the data recognition scenario.

[0153] wherein, the state of restoring the reduction physical system is expressed as:

[0154] ;

[0155] wherein, is the sparse coefficient, is the residual term of the sparse identification algorithm, is the regularization term of the sparse identification algorithm, and y is the real data.

[0156] S105, According to the target physical system and the variable data collected by the system, perform real - time state prediction on the data recognition scenario, and the state prediction includes behavior data prediction and trend data prediction.

[0157] Among them, variable data refers to various measurement values or observation values related to the target physical system. These data can be temperature, pressure, speed, concentration, etc., and are used to describe the state of the system. The data recognition scenario refers to the scenario or process of using the collected data to understand and analyze the behavior of the physical system, which usually involves data processing and analysis. Real-time state prediction refers to continuously predicting the future state of the physical system based on current and historical data. Behavioral data prediction refers to predicting the behavioral characteristics of the physical system, such as how the system responds to external changes or internal dynamics. Trend data prediction refers to predicting the long-term trend of the physical system over time, such as growth, decrease, or periodic changes.

[0158] In some embodiments, assume that a city has installed a large number of traffic monitoring cameras and sensors, and the city's traffic system collects data on traffic flow, vehicle speed, and traffic accidents. Using this "variable data", the city's traffic system can perform "data recognition scenario" analysis, such as identifying traffic congestion patterns and causes. Using the results of these analyses, the traffic management center can perform "real-time state prediction", such as predicting traffic flow and congestion during the evening rush hour. They can use machine learning algorithms to predict traffic conditions in the next few hours based on current traffic flow data and historical patterns.

[0159] For example, "behavioral data prediction" may involve predicting how vehicles will re-route when a major road is closed for construction and how this will affect traffic flow and congestion on surrounding roads.

[0160] For example, "trend data prediction" may include predicting which areas will require more traffic infrastructure as the city expands and the population grows, or predicting the long-term growth trend of public transportation demand.

[0161] By applying the technical solution of the present application, in the sparse identification algorithm of the nonlinear dynamic system used in restoring the physical system in a specific data recognition scenario, a method of constructing the related function B with the threshold λ as the independent variable can be added to adaptively select the threshold for the nonlinear dynamic system, so as to use the nonlinear dynamic system configured with the self-selected threshold to restore the state information of a physical system within a certain period of time. Thus, the purpose of correctly restoring the state of the physical system without manual threshold selection is achieved. It also solves the reduction error problem caused by missing function terms that may occur in the process of using the nonlinear dynamic system to restore the physical system in the related art, even when a relatively small threshold is selected.

[0162] In another embodiment of the present application, as Figure 4 shown, the present application also provides a state prediction device based on physical system state restoration. It includes:

[0163] A creation module 201 is configured to determine a physical system to be restored that exists in a real environment, and create a non-linear dynamic system for capturing non-linear characteristics of the physical system to be restored, where the non-linear characteristics are used to reflect the real behavior of the physical system to be restored;

[0164] A calculation module 202 is configured to calculate a first function and a second function for the non-linear dynamic system, where the first function is a functional relationship reflecting the data error between the simulated data and the real data of the physical system to be restored, and the second function is a functional relationship reflecting the sparsity between the threshold and the physical system to be restored, where the physical system to be restored is a physical system whose state needs to be restored, and the threshold is used to control the sparsity degree of the output value of the non-linear dynamic system;

[0165] A selection module 203 is configured to use the first function and the first function to calculate a correlation function for the non-linear dynamic system, and use the correlation function to select a corresponding target value for the non-linear dynamic system;

[0166] An analysis module 204 is configured to configure the target value as a self-selection threshold of the non-linear dynamic system, and use the non-linear dynamic system configured with the self-selection threshold to perform state restoration on the physical system to be restored in a data recognition scenario for a preset duration, to obtain a target physical system after state restoration;

[0167] A state prediction module 205 is configured to perform real-time state prediction on the data recognition scenario according to the target physical system and variable data collected by the system, where the state prediction includes behavior data prediction and trend data prediction.

[0168] By applying the technical solution of the present application, in the sparse identification algorithm of the non-linear dynamic system that can be used in the physical system to be restored in a certain data recognition scenario, a method of adaptively selecting a threshold for the non-linear dynamic system can be achieved by adding a method of constructing a correlation function B with the threshold λ as an independent variable, so that the non-linear dynamic system configured with the self-selection threshold is subsequently used to restore the state information of a certain physical system within a period of time. Thus, an object of correctly restoring the state of the physical system without manual threshold selection can be achieved. It also solves the problem of the reduction error caused by missing function terms that may still occur in the process of using the non-linear dynamic system to restore the physical system in the related art, even when a relatively small threshold is selected.

[0169] In another embodiment of the present application, the selection module 202 is configured as:

[0170] The first function is calculated for the non - linear dynamic system by the following formula:

[0171] L ;

[0172] wherein, the L is the first function, which is used to reflect the functional relationship between the real data and the fitted data.

[0173] In another embodiment of the present application, the selection module 202 is configured to:

[0174] The second function is calculated for the non - linear dynamic system by the following formula:

[0175] ;

[0176] wherein, the is the second function, the is the number of all non - zero terms in the function library of the reduced physical system, and the is the number of all column terms in the function library.

[0177] In another embodiment of the present application, the selection module 202 is configured to:

[0178] The correlation function is calculated for the non - linear dynamic system by the following formula:

[0179] ;

[0180] wherein, the is the correlation function, the L is the first function, and the is the second function.

[0181] In another embodiment of the present application, the selection module 202 is configured to:

[0182] Determine the value range corresponding to the maximum value range of the correlation function value in the correlation function;

[0183] Take the minimum value in the value range as the target value.

[0184] In another embodiment of the present application, the selection module 202 is configured to:

[0185] Take the target The value is used as the self - selection threshold of the sparse identification algorithm for the non - linear dynamic system;

[0186] Among them, the sparse identification algorithm is obtained from the following formula:

[0187] ;

[0188] Among them, is the sparse coefficient, the is the residual term of the sparse identification algorithm, the is the regularization term of the sparse identification algorithm, and y is the real data.

[0189] In another embodiment of the present application, the selection module 202 is configured to:

[0190] Use the non - linear dynamic system configured with the self - selection threshold to perform state restoration on the test physical system to obtain a test restoration result; and collect the test real result corresponding to the test physical system in the real situation;

[0191] When it is detected that the error value between the test restoration result and the test real result is less than a preset difference value, use the non - linear dynamic system configured with the self - selection threshold to perform state restoration on the restored physical system in the data identification scenario for a preset duration.

[0192] An embodiment of the present application also provides an electronic device to execute the state prediction method based on the state restoration of the physical system. Please refer to Figure 5 , which shows a schematic diagram of an electronic device provided by some embodiments of the present application. As Figure 5 shown, the electronic device 3 includes: a processor 300, a memory 301, a bus 302, and a communication interface 303. The processor 300, the communication interface 303, and the memory 301 are connected through the bus 302; a computer program that can run on the processor 300 is stored in the memory 301, and when the processor 300 runs the computer program, it executes the state prediction method based on the state restoration of the physical system provided by any of the foregoing embodiments of the present application.

[0193] Among them, the memory 301 may include a high - speed random access memory (RAM: Random Access Memory), and may also include a non - volatile memory, such as at least one disk memory. Through at least one communication interface 303 (which can be wired or wireless), communication connection between this device network element and at least one other network element can be realized, and the Internet, wide - area network, local area network, metropolitan area network, etc. can be used.

[0194] The bus 302 can be an ISA bus, a PCI bus, an EISA bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, etc. Among them, the memory 301 is used to store a program. After receiving an execution instruction, the processor 300 executes the program. Any implementation manner of the video transmission method disclosed in any implementation manner of the embodiments of the present application can be applied to the processor 300 or implemented by the processor 300.

[0195] The processor 300 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit in the hardware of the processor 300 or by instructions in software form. The above-mentioned processor 300 can be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), 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, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory 301, and the processor 300 reads the information in the memory 301 and combines its hardware to complete the steps of the above method.

[0196] The electronic device provided in the embodiments of the present application and the state prediction method based on physical system state restoration provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run, or implemented by them.

[0197] The embodiments of the present application also provide a computer-readable storage medium corresponding to the state prediction method based on physical system state restoration provided in the foregoing embodiments. Please refer to Figure 6 which shows that the computer-readable storage medium is an optical disc 40, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the method provided in any of the foregoing embodiments.

[0198] It should be noted that examples of the computer-readable storage medium may further include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, or other optical and magnetic storage media, which will not be elaborated herein one by one.

[0199] The computer-readable storage medium provided by the above embodiments of the present application and the video transmission method provided by the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run, or implemented by the application programs stored therein.

[0200] It should be noted that:

[0201] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that the embodiments of the present application may be practiced without these specific details. In some instances, well-known structures and technologies have not been shown in detail so as not to obscure the understanding of this specification.

[0202] Similarly, it should be understood that, in order to streamline the present application and assist in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present application, the various features of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting the following schematic: that the claimed present application requires more features than those expressly recited in each claim. Rather, as reflected in the following claims, the inventive aspects lie in less than all the features of the single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into the detailed description, where each claim stands on its own as a separate embodiment of the present application.

[0203] In addition, those skilled in the art can understand that, although some of the embodiments described herein include certain features included in other embodiments but not other features, the combination of the features of different embodiments means that it is within the scope of the present application and forms different embodiments. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

[0204] The above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A state prediction method based on physical system state restoration, characterized in that: The method comprises: Determine a traffic restoration physical system to be restored in a real environment, and create a nonlinear dynamic system for capturing the nonlinear characteristics of the traffic restoration physical system, wherein the nonlinear characteristics are used to reflect the real behavior of the traffic restoration physical system; The first function and the second function are calculated for the nonlinear dynamic system, wherein the first function is used to reflect the functional relationship between the data error between the simulated data and the real data of the traffic restoration physical system, and the second function is used to reflect the threshold and the functional relationship between the sparsity of the traffic restoration physical system, the traffic restoration physical system is a physical system that needs to be restored to its state, and the threshold Used to control the sparsity of the output value of the nonlinear dynamic system; Using the first function and the first function, a correlation function is calculated for the nonlinear dynamic system, and the correlation function is used to select a corresponding target for the nonlinear dynamic system. value; The correlation function is calculated for the nonlinear dynamic system by the following formula: ； Among them, the is the correlation function, the L is the first function, the is the second function; The method uses the correlation function to select a corresponding target for the nonlinear dynamic system Values, including: Determine in the correlation function, the value of the correlation function corresponding to the maximum value interval Value interval; The The minimum value in the value interval is taken as the target value; The goal The value is configured as a self-selected threshold of the nonlinear dynamic system, and the nonlinear dynamic system configured with the self-selected threshold is used to perform state restoration within a preset time period on the traffic restoration physical system in the data recognition scenario to obtain a target physical system after state restoration; According to the target physical system and variable data collected by the system, the variable data are vehicle flow, vehicle speed and traffic accident data, real-time state prediction is performed on the data recognition scene, and the state prediction includes behavior data prediction and trend data prediction; Among them, behavioral data prediction includes predicting the routes chosen by vehicles when major roads are closed for construction, as well as the impact on traffic and congestion on surrounding roads; Trend data forecasts include predicting long-term growth trends in public transport demand.

2. The method according to claim 1, characterized in that The first function is calculated for the nonlinear dynamic system by the following formula: L ; Among them, the L is the first function, Used to reflect the functional relationship between real data and fitting data.

3. The method according to claim 2, characterized in that The second function is calculated for the nonlinear dynamic system by the following formula: ； Among them, the is the second function, The number of all non-zero entries in the function library in the traffic restoration physical system, the is the number of all columns in the function library.

4. The method according to claim 1, characterized in that The target The values ​​are configured as self-selected thresholds for the nonlinear dynamic system and include: The goal The value is used as a self-selected threshold of the sparse identification algorithm of the nonlinear dynamic system; The sparse recognition algorithm is obtained by the following formula: ； in, is the sparse coefficient, is the residual term of the sparse recognition algorithm, is the regularization term of the sparse recognition algorithm, y is the real data, and A is the basic function library.

5. The method according to claim 1, characterized in that The method of using the nonlinear dynamic system configured with the self-selected threshold to restore the state of the traffic restoration physical system in the data recognition scenario within a preset time period includes: Performing state restoration on the test physical system using the nonlinear dynamic system configured with the self-selected threshold to obtain a test restoration result; and collecting a real test result corresponding to the test physical system under real conditions; When it is detected that the error value between the test restoration result and the actual test result is less than the preset difference, the nonlinear dynamic system configured with the self-selected threshold is used to restore the state of the traffic restoration physical system in the data recognition scenario within a preset time period.

6. A state prediction device based on physical system state restoration implemented by the method described in any one of claims 1 to 5, characterized in that: The device comprises: A creation module is used to determine a traffic restoration physical system to be restored in a real environment, and to create a nonlinear dynamic system for capturing the nonlinear characteristics of the traffic restoration physical system, wherein the nonlinear characteristics are used to reflect the real behavior of the traffic restoration physical system; The calculation module is configured to calculate the first function and the second function by the nonlinear dynamic system, wherein the first function is used to reflect the functional relationship of the data error between the simulated data and the real data of the traffic restoration physical system, and the second function is used to reflect the threshold and the functional relationship between the sparsity of the traffic restoration physical system, the traffic restoration physical system is a physical system that needs to be restored to its state, and the threshold Used to control the sparsity of the output value of the nonlinear dynamic system; A selection module is configured to use the first function and the first function to calculate a correlation function for the nonlinear dynamic system, and use the correlation function to select a corresponding target for the nonlinear dynamic system value; The analysis module is configured to convert the target The value is configured as a self-selected threshold of the nonlinear dynamic system, and the nonlinear dynamic system configured with the self-selected threshold is used to perform state restoration within a preset time period on the traffic restoration physical system in the data recognition scenario to obtain a target physical system after state restoration; The state prediction module is used to perform real-time state prediction on the data recognition scenario based on the target physical system and the variable data collected by the system, wherein the state prediction includes behavior data prediction and trend data prediction.

7. An electronic device, characterized in that: include: A memory for storing executable instructions; as well as, A processor is used to execute the executable instructions with the memory to complete the operation of the state prediction method based on physical system state restoration as described in any one of claims 1-5.

8. A computing device readable storage medium for storing instructions readable by a computing device, characterized in that: When the instruction is executed, the operation of the state prediction method based on physical system state restoration described in any one of claims 1-5 is performed.

Images