Clean environment intelligent logistics management system based on digital twin

Abstract

The present invention relates to the field of intelligent environmental management and the technical field of the Internet of Things, and further to a clean environment intelligent logistics management system based on digital twins. The system comprises: a clean environment intelligent logistics management model establishment unit, which is used to regard the clean environment intelligent logistics management system as an evolutionary system with a continuous time state and establish a corresponding abstract model; a digital twin management and prediction unit, which is used to establish a corresponding observation equation according to the established abstract model, perform continuous time Kalman filtering, perform continuous time parameter estimation, and then perform continuous time model optimization, and finally discretize the continuous time to obtain a discrete time state equation, and control the clean environment intelligent logistics management system based on the estimated discrete state transfer matrix and discrete control input matrix. The present invention realizes high-precision control and real-time response of the clean environment through digital twin technology, reduces energy consumption and maintenance costs, and improves the quality and efficiency of environmental management.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent environmental management and the technical field of the Internet of Things, and specifically relates to a clean environment intelligent logistics management system based on digital twins. Background Art

[0002] Clean environment management is a vital task in many fields, including medical, pharmaceutical, semiconductor production, laboratory research, etc. In these fields, precise control and monitoring of environmental parameters such as temperature, humidity, pressure difference, air quality, light brightness and noise level are essential to ensure a clean, stable and safe working environment. Traditional clean environment management systems mainly rely on timing, manual operation, or use simple feedback control methods, which are often difficult to cope with rapidly changing environmental conditions and complex control requirements.

[0003] Traditional control methods have difficulty in achieving high-precision control of environmental parameters. For example, the control of temperature and humidity may be interfered with by external factors, causing parameter fluctuations beyond the allowable range. This affects many sensitive applications, such as laboratory research and medical surgery. Traditional systems have limited real-time response capabilities and are difficult to adapt to environmental changes quickly. When environmental parameters change suddenly, the system may take a long time to adjust, resulting in instability and inadaptability of the working environment. Existing systems often have difficulty effectively handling sources of uncertainty and noise, such as sensor errors, external interference, etc. These uncertainties can have a negative impact on the performance of the control system. Some traditional systems may maintain high energy consumption when it is not necessary to ensure the constancy of environmental parameters. This approach not only wastes energy, but may also increase operating costs. Traditional clean environment management systems usually require frequent manual maintenance and calibration, which increases maintenance costs and system downtime.

[0004] In order to solve these problems, some advanced technologies and methods have been proposed and applied in the field of clean environment management in recent years. One of the important technologies is digital twin technology, which allows the establishment of a digital model of the clean environment intelligent logistics management system, and improves the performance and efficiency of the system through real-time data acquisition, state estimation and control strategy optimization. The key concept of digital twin technology is to match the actual system with its digital model to achieve real-time monitoring and control. In this technology, the clean environment intelligent logistics management system is regarded as an evolving system of continuous time state, whose state consists of environmental quality parameters, control input parameters, system response actions and control input actions. These parameters and actions are abstracted into state vectors, control input vectors, state transfer matrices and control input matrices to form an abstract model. Digital twin technology also includes the use of continuous time Kalman filtering to estimate state vectors and parameters and perform model optimization. Continuous time Kalman filtering uses observation equations and state equations to update state estimates, while considering the effects of observation noise and process noise. This helps the system to monitor environmental parameters, control systems and cope with uncertainties in real time. Summary of the invention

[0005] The main purpose of the present invention is to provide a clean environment intelligent logistics management system based on digital twins. The present invention uses digital twin technology to achieve high-precision control and real-time response of the clean environment, reduce energy consumption and maintenance costs, and improve the quality and efficiency of environmental management.

[0006] In order to solve the above problems, the technical solution of the present invention is achieved as follows:

[0007] A clean environment intelligent logistics management system based on digital twins, the system includes: a clean environment intelligent logistics management model establishment unit, which is used to regard the clean environment intelligent logistics management system as an evolving system with a continuous time state and establish a corresponding abstract model, specifically including: within a set time period, collecting the operating parameters of the clean environment intelligent logistics management system under continuous time, the operating parameters include: environmental quality parameters, control input parameters, control input actions, system response actions and noise; the environmental quality parameters are used as the state vector of the abstract model; the control input parameters are defined as the control parameter set issued by the clean environment intelligent logistics management system at a certain moment, and the control parameters are used as the control input vector of the abstract model; the system response action is defined as the clean environment intelligent logistics management system executing the corresponding action according to the data of the external environment perceived at a certain moment. The system response action is a state transfer matrix of the abstract model; the control input action is defined as the action set received by the clean environment intelligent logistics management system at a certain moment, and the control input action is the control input matrix of the abstract model; the digital twin management and prediction unit is used to establish the corresponding observation equation according to the established abstract model, perform continuous-time Kalman filtering, perform continuous-time parameter estimation, and then perform continuous-time model optimization, and finally discretize the continuous time to obtain the discrete-time state equation, discretize the observation equation, use discrete-time Kalman filtering to estimate the discrete optimal state vector, and then estimate the discrete state transfer matrix and discrete control input matrix, and control the clean environment intelligent logistics management system based on the estimated discrete state transfer matrix and discrete control input matrix.

[0008] Furthermore, the environmental quality parameters include at least: temperature, humidity, pressure difference, air quality, light brightness and noise level; the control input parameters include at least: temperature setting value, humidity setting value, wind speed, wind direction and light brightness setting value.

[0009] Furthermore, the control input actions include at least: temperature increase, temperature decrease, humidity increase, humidity decrease, wind speed adjustment, wind direction adjustment, light brightness increase, light brightness decrease, air purification device turned on and air purification device turned off; the system response actions include at least: temperature increase, temperature decrease, humidity increase, humidity decrease, wind speed adjustment, wind direction adjustment, light brightness increase, light brightness decrease, air purification device turned on and air purification device turned off.

[0010] Furthermore, the abstract model is expressed by the following formula:

[0011] ;

[0012] in, is the abstract model variable, is the state vector; is the state transition matrix; is the control input matrix; is the control input vector; Represents the state vector Laplace operation of ; represents the control input integrated over time, which represents the cumulative effect of the control input over time; is the integration variable; For time The control input vector at time ; For time The control input matrix when ; is noise, representing nonlinear and time-varying system disturbances.

[0013] Furthermore, the observation equation is expressed using the following formula:

[0014] ;

[0015] in, is the observation vector; is the observation matrix; is the observation noise.

[0016] Furthermore, the following formula is used to perform continuous time Kalman filtering to obtain the optimal state estimation vector:

[0017] x ^ (t|t)= x ^ (t|t - 1)+K(t)[y(t)-C(t) x ^ (t|t - 1)] ;

[0018] in, is the optimal state estimate vector; is the optimal state estimation vector at the previous moment; is the Kalman gain.

[0019] Furthermore, the method for continuous-time parameter estimation includes: defining a state transfer matrix and the control input matrix The loss function is: By minimizing the loss function, the optimization problem of parameter estimation is obtained; by solving the optimization problem of parameter estimation, the state transfer matrix and the control input matrix are estimated; the loss function is expressed as:

[0020] ;

[0021] in, is the loss function value; by minimizing the loss function, the optimization problem of parameter estimation is expressed using the following formula:

[0022] ;

[0023] is the estimated state transfer matrix; To estimate the control input matrix, use the following formula to solve the optimization problem of parameter estimation:

[0024]

[0025] based on and The continuous time model is optimized, and the optimized abstract model is:

[0026] .

[0027] Furthermore, the optimized abstract model is discretized in continuous time, and the discrete time state equation is expressed using the following formula:

[0028] ;

[0029] in, is a discrete time variable; is a discrete integral variable.

[0030] Furthermore, the observation equation is discretized by the following formula:

[0031] y[k]=C( k ) x ̇ (k)+v(k) ;

[0032] The discrete time Kalman filter is used to estimate the discrete optimal state vector by the following formula: x ^ [k|k] :

[0033] x ^ [k|k]= x ^ [k|k - 1]+ A ^ k K[k][y[k]- B ^ k P[k|k - 1] P k k = x ^ k|k - 1 + A ^ (k)P[k - 1|k - 1] B ^ T (k)+Q[k] ;

[0034] in, is the discrete state covariance matrix; Q[k] is the discrete process noise covariance matrix; K[k] is the discrete Kalman gain; then according to the estimated discrete optimal state vector x ^ [k|k] Estimate the discrete state transfer matrix and discrete control input matrix.

[0035] The clean environment intelligent logistics management system based on digital twin of the present invention has the following beneficial effects: traditional clean environment management systems are usually difficult to achieve high-precision control of environmental parameters. The present invention adopts digital twin technology, and through methods such as continuous time Kalman filtering, environmental parameters such as temperature, humidity, brightness, etc. can be estimated more accurately. This means that the system can monitor and adjust environmental parameters in real time to ensure that they are always within a predetermined range. For example, in medical surgery, the control of temperature and humidity is crucial. The present invention can ensure that the environmental parameters in the operating room always meet medical standards, thereby improving the success rate of surgery and the safety of patients. The clean environment management system of the present invention has excellent real-time response capabilities. Through digital twin technology, the system can quickly collect real-time data and perform state estimation, and then perform corresponding control operations according to the estimated value. This means that the system can quickly adapt to rapidly changing environmental conditions, thereby improving the stability and controllability of the working environment. In fields such as semiconductor manufacturing, this real-time response capability is crucial to ensuring product quality. Existing clean environment management systems are often interfered by uncertainty and noise, such as sensor errors, external interference, etc. The present invention can effectively handle these uncertain factors through technologies such as continuous time Kalman filtering. Kalman filtering can not only estimate the state vector, but also estimate the state covariance matrix, providing quantitative information about uncertainty. This enables the system to better adapt to unstable environmental conditions and improves the robustness and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic diagram of the system structure of a clean environment intelligent logistics management system based on digital twins provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0038] The following are detailed descriptions of each.

[0039] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that shown or described herein.

[0040] Example 1: Reference Figure 1 , a clean environment intelligent logistics management system based on digital twin, the system includes: a clean environment intelligent logistics management model establishment unit, which is used to regard the clean environment intelligent logistics management system as an evolutionary system with continuous time state, and establish a corresponding abstract model, specifically including: within a set time period, collecting the operating parameters of the clean environment intelligent logistics management system under continuous time, the operating parameters include: environmental quality parameters, control input parameters, control input actions, system response actions and noise; the environmental quality parameters are used as the state vector of the abstract model; the control input parameters are defined as the control parameter set issued by the clean environment intelligent logistics management system at a certain moment, and the control parameters are used as the control input vector of the abstract model; the system response action is defined as the clean environment intelligent logistics management system at a certain moment according to the data of the external environment perceived, and the execution An action set of corresponding actions, each element in the action set is composed of a numerical code corresponding to the action, and the system response action is used as the state transfer matrix of the abstract model; the control input action is defined as the action set received by the clean environment intelligent logistics management system at a certain moment when the user input is received, and the control input action is used as the control input matrix of the abstract model; the digital twin management and prediction unit is used to establish the corresponding observation equation according to the established abstract model, perform continuous-time Kalman filtering, perform continuous-time parameter estimation, and then perform continuous-time model optimization, and finally discretize the continuous time to obtain the discrete-time state equation, discretize the observation equation, use discrete-time Kalman filtering to estimate the discrete optimal state vector, and then estimate the discrete state transfer matrix and discrete control input matrix, and control the clean environment intelligent logistics management system based on the estimated discrete state transfer matrix and discrete control input matrix.

[0041] Specifically, the clean environment is regarded as a continuous-time state evolution system, and an abstract model is established to monitor, respond and optimize the cleanliness of the environment in real time. The system continuously collects environmental quality parameters, external control input parameters, user input actions and system response actions, and establishes a mathematical model based on this information. Then, continuous-time Kalman filtering and parameter estimation techniques are used to optimize the model and discretize it for actual control.

[0042] The state vector is the key environmental quality parameter in the system, such as temperature, humidity, air quality, etc. These parameters reflect the actual condition of the environment in the clean environment intelligent logistics management. By tracking the state vector, the system can understand the current state of the environment. This is critical to maintaining a clean environment because the quality of the environment directly affects the success and efficiency of production, experiments, or other operations. The control input vector contains a set of external control parameters received by the system, which can be used to adjust the operation of the clean environment intelligent logistics management system. Through the control input vector, the operator can intervene in the system in real time to meet specific requirements or goals. This can include adjusting temperature, humidity, or other parameters to maintain suitable conditions for the clean environment. The state transition matrix describes how the system responds to changes in the external environment. It contains a set of actions that the system performs in different situations. The state transition matrix tells the system how to perform actions based on the perceived external environmental data to adapt to or improve the quality of the environment. For example, if the temperature rises, the state transition matrix can include an action to reduce the power of the cooling system. The control input matrix contains a set of actions received by the system as user input. These actions can affect the operation of the system. The control input matrix allows operators or automated programs to issue commands to achieve specific control goals. For example, an operator can send instructions to start or stop a specific device, or to increase / decrease the frequency of a specific operation.

[0043] The state vector in the abstract model, i.e., the environmental quality parameters, acts as a real-time monitor of the system's clean environment. By continuously monitoring these parameters, the system is able to sense changes in the environment and take action when necessary. For example, if the temperature or humidity exceeds the set range, the system can react immediately to prevent problems that may affect the production process. The control input vector and the control input matrix are key tools for implementing control strategies. When the clean environment needs to be adjusted, the operator or automation program can send appropriate instructions to the system through the control input matrix. These instructions can include changing the operating state of the equipment, adjusting parameter settings, or performing other operations to maintain or optimize environmental quality. The state transition matrix of the abstract model reflects how the system responds to changes in the external environment. This allows the system to self-optimize based on real-time monitoring. For example, if the state transition matrix indicates that performing certain actions under specific conditions can improve environmental quality, the system can automatically adjust its operation to achieve more efficient operation. This helps improve resource utilization and reduce unnecessary energy consumption. The various elements in the abstract model can also be used for data recording and analysis. Historical data of the state vector can be used to generate environmental quality trend charts, while the control input matrix and the state transition matrix can be used to analyze the performance of the system under different conditions. This data helps managers make more informed decisions and improve the operation of the clean environment smart logistics management system. This abstract model provides a comprehensive framework for clean environment smart logistics management, enabling the system to respond to environmental changes in real time, execute control strategies, self-optimize, and provide data support to improve the efficiency, reliability, and controllability of the clean environment. This integrated approach helps ensure the stability of the clean environment, especially in applications with strict requirements on environmental quality, such as medical device manufacturing, life science laboratories, and other fields.

[0044] The digital twin management and prediction unit uses continuous-time Kalman filtering technology to estimate the continuous-time state of the system. This means that it can monitor the state of the system in real time, including environmental quality parameters and other key parameters. This real-time monitoring helps to detect any anomalies or changes in a timely manner. Through continuous-time Kalman filtering, the system is able to continuously optimize the state vector and parameter estimates in the abstract model. This means that the system can adapt to dynamic environments and continuously improve performance. For example, if the state vector of the system indicates that the environmental quality has deteriorated, the system can automatically take measures to correct the situation to maintain or improve the clean environment. Continuous-time Kalman filtering not only uses information from abstract models, but also combines actual observation data to more accurately estimate the state of the system. This fusion can reduce the gap between the model and the real environment, making the system's predictions more accurate. Based on real-time state estimates, the digital twin management and prediction unit can provide information about the behavior of the system to help operators or automated systems make immediate decisions. This can include adjusting control inputs, triggering alarms, or taking other actions to ensure the suitability and stability of the clean environment.

[0045] The continuous-time state equation describes how the system state evolves over time. Discretization converts this continuous-time equation into a state update process at discrete time steps, usually approximated using difference equations such as the Euler method or the Runge-Kutta method. This allows the system state to be estimated at discrete time points. The observation equation describes how the system state is related to the observation data. The discretized observation equation maps the relationship between the actual observation data and the model state to discrete time points in order to combine the observations with the state estimate. After discretizing the continuous-time equation and the observation equation, the system can apply the discrete-time Kalman filter algorithm. The algorithm recursively updates the estimate of the system state based on the estimated state vector and the observation data. This allows the system to estimate the state at discrete time points and make state predictions and updates in continuous time. The discrete state transition matrix describes how the system transitions from one state to the next at discrete time steps, while the discrete control input matrix represents how the control input is applied at each time step to affect the state change. Based on the estimated discrete state transition matrix and discrete control input matrix, the system can perform actual control actions to maintain or optimize the clean environment. This can include adjusting environmental parameters such as air conditioning, ventilation systems, or lighting to keep environmental quality within expected ranges. Control systems can perform responsive actions that are the result of decisions based on system status and control inputs. Response actions may include alarm triggering, equipment starting or stopping, and other operations in response to environmental changes or to meet preset goals.

[0046] Embodiment 2: The environmental quality parameters include at least: temperature, humidity, pressure difference, air quality, light brightness and noise level; the control input parameters include at least: temperature setting value, humidity setting value, wind speed, wind direction and light brightness setting value.

[0047] Temperature is the heat level of the environment, usually expressed in degrees Celsius or Fahrenheit. In clean environments, temperature control is very important to ensure that the temperature conditions required for different applications are maintained. Humidity refers to the amount of water vapor in the air. In clean environments, humidity maintenance is critical for some applications such as laboratories or production processes because it can affect the deposition and propagation of particulate matter in the air. Pressure differential refers to the gas pressure in the environment. It is usually expressed in Pascals (Pa) or hectopascals (hPa). In some special applications, such as aerospace or high altitude areas, the control of pressure differential may be important. Air quality includes information about pollutants such as particulate matter, gases, and chemicals in the air. Clean environments need to maintain a high level of air quality to reduce the impact of harmful substances on personnel health and equipment performance. Brightness refers to the light intensity level in the environment, usually expressed in lux. The control of light brightness is very important for lighting and visual tasks, especially in places such as laboratories, medical facilities, and offices. Noise level refers to the sound intensity level in the environment, usually expressed in decibels (dB). In some applications, such as laboratories or medical facilities, the control of noise level is critical to ensure a normal working and research environment.

[0048] The temperature setpoint is the desired temperature level set by the operator or automated system. The system will try to maintain the ambient temperature close to this setpoint. The humidity setpoint is the desired humidity level, usually expressed as a percentage of relative humidity. The system will strive to keep the humidity close to the setpoint. Wind speed is the speed of air flow and can be controlled by adjusting the fan or ventilation system. It is very important for improving air quality and temperature distribution. Wind direction indicates the direction of air flow. Control of wind direction can ensure that gases and particulate matter in the clean environment are evenly distributed. The light brightness setpoint is the desired light level. The system can achieve this setpoint by adjusting the lighting equipment. The comprehensive control of these environmental quality parameters and control input parameters can ensure that the clean environment maintains the required conditions in different application scenarios, thereby ensuring the success and efficiency of production, experiments or other operations. By monitoring and adjusting these parameters in real time, the clean environment intelligent logistics management system can achieve a high degree of controllability and automated management.

[0049] Embodiment 3: The control input actions include at least: temperature increase, temperature decrease, humidity increase, humidity decrease, wind speed adjustment, wind direction adjustment, light brightness increase, light brightness decrease, air purification device on and air purification device off; the system response actions include at least: temperature increase, temperature decrease, humidity increase, humidity decrease, wind speed adjustment, wind direction adjustment, light brightness increase, light brightness decrease, air purification device on and air purification device off.

[0050] Temperature increase: Increase the ambient temperature by increasing the power of the heating device or adjusting the HVAC system. Temperature decrease: Decrease the ambient temperature by reducing the power of the heating device or adjusting the cooling device. Humidity increase: Increase the ambient humidity by increasing the humidity output of the humidifier. Humidity decrease: Decrease the ambient humidity by decreasing the humidity output of the humidifier. Wind speed adjustment: Change the intensity of air movement by controlling the speed of the fan or ventilation system. Wind direction adjustment: Change the direction of air movement by adjusting the wind direction of the ventilation system. Light brightness increase: Increase the ambient light level by increasing the brightness of the lighting equipment. Light brightness decrease: Decrease the ambient light level by decreasing the brightness of the lighting equipment. Air purification device on: Activate the air purification device to improve air quality and reduce pollutants and particulate matter. Air purification device off: Turn off the air purification device to stop the purification process.

[0051] Embodiment 4: The abstract model is represented by the following formula:

[0052] ;

[0053] in, is the abstract model variable, is the state vector; is the state transition matrix; is the control input matrix; is the control input vector; Represents the state vector Laplace operation of ; represents the control input integrated over time, which represents the cumulative effect of the control input over time; is the integration variable; For time The control input vector at time ; For time The control input matrix when ; is noise, representing nonlinear and time-varying system disturbances.

[0054] Specifically, Represents the system state vector Over time In clean environment management, the state vector can include environmental parameters such as temperature, humidity, and brightness. It reflects how these parameters change over time and is the basis for monitoring environmental changes. Describes the dynamic evolution of state vectors, that is, how they influence each other and change over time. In clean environment management, The interactions between environmental parameters can be represented, such as the coupling relationship between temperature and humidity. Indicates external control input How to influence the state change of the system. In clean environment management, It can be used to model control operations, such as adjusting temperature, humidity, or light levels, to maintain or optimize environmental quality. Represents the state vector Laplace operation is used to describe the spatial variation between state variables. In clean environment management, it can be used to model processes such as heat transfer, mass transfer and flow in the environment, which may affect the distribution of state variables in practical applications. Indicates at time Internal control input The integration shows the cumulative effect of the control input over time. In clean environment management, this can be used to consider the impact of control operations over a period of time on the environmental state, such as the cumulative effect of the state after multiple control operations. Represents the nonlinear and time-varying disturbances in the system, which may come from external interference, sensor errors, or internal uncertainties in the system. In clean environment management, the noise term can be used to model uncertainty and randomness in the environment. In summary, the principle of this formula is to abstract the dynamic behavior of the clean environment intelligent logistics management system into a mathematical model. By analyzing this model, we can understand how the system state is affected by external control, spatial variation, and uncertainty, so as to optimize the control strategy to maintain or improve the quality and stability of the clean environment. This model can be used to simulate and predict the behavior of the system, providing a basis for system operation and management.

[0055] Embodiment 5: The observation equation is expressed using the following formula:

[0056] ;

[0057] in, is the observation vector; is the observation matrix; is the observation noise.

[0058] Specifically, Represents the observation vector of the system, which contains a set of observed variables or measurement results. In clean environment management, the observation vector can include the measured values ​​of various environmental parameters, such as temperature, humidity, brightness, etc. is the observation matrix, which describes the state vector How to map to observation vector The observation matrix represents the relationship between the observation value and the system state. In clean environment management, It can represent the characteristics and location of sensors, as well as the relationship between observed variables and state variables. is the observation noise term, which represents the random disturbance and error in the observation value. The observation noise can come from sensor error, environmental interference or measurement uncertainty. In clean environment management, It can be used to model the uncertainty in the observation process. The observation equation describes the system state vector How to use the observation matrix Mapping to observation vector , and taking into account the observation noise This equation is often used to relate actual observations to system states to monitor changes and variations in environmental parameters. Through the observation equation, the system can use sensors or measuring devices to obtain information about environmental parameters in real time in order to implement feedback control and state estimation to maintain or optimize the quality of the clean environment. The observation equation plays a key role in system monitoring and management, allowing the system to make decisions and adjustments based on actual observations.

[0059] Embodiment 6: Use the following formula to perform continuous time Kalman filtering to obtain the optimal state estimation vector:

[0060] x ^ (t|t)= x ^ (t|t - 1)+K(t)[y(t)-C(t) x ^ (t|t - 1)] ;

[0061] in, is the optimal state estimate vector; is the optimal state estimation vector at the previous moment; is the Kalman gain.

[0062] Specifically, Indicates at the current moment It is the best estimate of the system state, taking into account the observed value at the current moment. and the previous estimate . This estimate is based on a combination of past information and current observations. Indicates at time The optimal state estimate under . It is the state estimate obtained based on previous information and control input prediction. Kalman gain is a key parameter that is used to adjust the trade-off between observed and predicted values. The calculation of Kalman gain depends on the dynamic model and observation model of the system. It reflects the relationship between the credibility of the observed value and the credibility of the estimate. In clean environment management, Kalman gain can be used to effectively update state estimates while considering sensor accuracy and model uncertainty. Indicates at the current moment The actual observed value under the environment. In the clean environment management system, this can be environmental parameters measured by sensors, such as temperature, humidity, light brightness, etc. The goal of the clean environment management system is to monitor and maintain the cleanliness and comfort of the environment in real time. In order to do this, the system needs to estimate the state of the environment, such as temperature, humidity, etc., and perform corresponding control operations based on these states to maintain or optimize the environmental quality. Represents the best estimate of the environmental parameters at the current moment, taking into account past information and current observations. This estimate allows the system to better understand the current environmental conditions. Kalman Gain It plays a key role in providing an adjustment between the trade-off between observations and estimates. If the accuracy of the observations is high, the Kalman gain will adopt more observations, and vice versa. This helps the system make accurate estimates in an uncertain environment. By continuously updating , the system can adjust the control strategy based on the latest observation information, such as adjusting temperature, humidity or light brightness to ensure that the environment meets the set standards. This helps to improve energy efficiency, provide a comfortable working environment and maintain cleanliness. In summary, the principle of continuous-time Kalman filtering allows the clean environment intelligent logistics management system to estimate and maintain the optimal state of environmental parameters in real time by fusing observations and previous estimates, thereby achieving efficient environmental management and control.

[0063] Embodiment 7: The method for continuous time parameter estimation includes: defining a state transfer matrix and the control input matrix The loss function is: By minimizing the loss function, the optimization problem of parameter estimation is obtained; by solving the optimization problem of parameter estimation, the state transfer matrix and the control input matrix are estimated; the loss function is expressed as:

[0064] ;

[0065] in, is the loss function value; by minimizing the loss function, the optimization problem of parameter estimation is expressed using the following formula:

[0066] ;

[0067] is the estimated state transfer matrix; To estimate the control input matrix, use the following formula to solve the optimization problem of parameter estimation:

[0068]

[0069] based on and The continuous time model is optimized, and the optimized abstract model is:

[0070] .

[0071] Specifically, the loss function is a key metric used to measure the estimated state transition matrix and the control input matrix The error between the estimated value and the actual observed value. It quantifies this error by calculating the sum of the squares of the differences between the estimated value and the actual observed value. The goal of the loss function is to minimize this difference to obtain the best parameter estimate. In clean environment management, this means that the system tries to make its estimated state transfer matrix and control input matrix more consistent with the dynamic behavior of the real environment in order to better control and manage the environment. Minimizing the loss function It is an optimization process in which the system seeks to find the optimal state transfer matrix given the observed data. and the control input matrix By finding the parameter estimates that minimize the loss function, the system can make its model better match the way the environment operates. This is very important for ensuring the accuracy of the system's state estimation and control strategy, especially in complex and changing environmental conditions. Once the loss function is minimized , the system can obtain an estimated value of the state transfer matrix and estimates of the control input matrix These estimates reflect the dynamic behavior of the environmental system and the impact of control inputs. In clean environment management, these estimates can be used to improve environmental control strategies, such as adjusting temperature, humidity, light brightness, etc. to meet set requirements and standards. and the control input matrix , the system can obtain an optimized abstract model. This model includes the continuous-time state equation, which takes into account the state transfer matrix, control input matrix, state space change and noise terms. By optimizing the model, the system can more accurately predict the behavior of the environment and implement more effective control strategies to maintain a clean environment and provide a comfortable working environment.

[0072] Embodiment 8: The optimized abstract model is discretized in continuous time to obtain a discrete time state equation which is expressed using the following formula:

[0073] ;

[0074] in, is a discrete time variable; is a discrete integral variable.

[0075] Specifically, the goal of the clean environment intelligent logistics management system is to monitor and control environmental parameters, such as temperature, humidity, brightness, etc., to maintain a clean and comfortable working environment. In order to achieve this goal, the system needs to establish a mathematical model to describe the changes in environmental parameters and the impact of control operations. The initial system modeling usually adopts a continuous-time model, which takes into account the continuous evolution of environmental parameters, the continuous application of control operations, and possible noise and nonlinear factors. Such a model can more accurately capture the dynamic behavior of the environment. Although the continuous-time model is very powerful, in the actual control and estimation process, it is usually necessary to discretize the model to adapt to the needs of computer control and data processing. In addition, many sensors and control systems work in discrete time. Therefore, it is necessary to convert the continuous-time model into a discrete-time model. The above formula expresses the result of this conversion. Among them, x ˙ [ k +1] Represents the discrete time step Lower state vector The state vector includes environmental parameters such as temperature and humidity. and is a discrete time step The estimated values ​​of the state transfer matrix and control input matrix under , which describe the evolution of the state and the impact of the control input. The spatial variation of states is taken into account to better describe gradients in the environment. The cumulative effect of control inputs over discrete time intervals is considered to better understand the impact of control actions. Ω k ] Represents discrete time steps The noise term under is used to account for system disturbances. This discrete-time state equation design is suitable for clean environment intelligent logistics management systems because it allows the system to perform state estimation and control operations within discrete time steps. The system can monitor the environment in real time, adjust the control strategy, and cope with noise and uncertainty based on the state change rate, control input, and spatial changes in discrete time steps.

[0076] Discrete-time state variables x ˙ [ k +1] In discrete time steps Lower state vector In clean environment management, the state vector This can include environmental parameters such as temperature, humidity, light intensity, etc. By tracking the rate of change of the state, the system can understand how the environmental parameters evolve over time. and the control input matrix Used to describe the system in discrete time steps arrive The state evolution between and the influence of control input. In the clean environment management system, may represent the natural evolution trend of environmental parameters, while May represent the effect of a control operation on an environmental parameter, such as adjusting temperature or humidity. Considering the state vector The spatial variation of the state, that is, the gradient of the state in space. In clean environment management, this may represent the distribution of environmental parameters in physical space, such as temperature gradient. In discrete time steps arrive Control input between The integral of is a discrete integral variable. This takes into account the cumulative effect of a control action over a period of time. In clean environment management, this can represent the cumulative impact of a control action on an environmental parameter, such as the effect of heating or reducing humidity. Ω k ] Represents the discrete time step The noise term under is used to account for nonlinear and time-varying disturbances in the system. In clean environment management, this can include environmental disturbances due to external factors (such as sudden temperature changes) or modeling errors. Including discrete time steps It can be used to take into account the influence of internal system dynamics or other factors to more accurately model the evolution of the state.

[0077] Embodiment 9: The observation equation is discretized by the following formula:

[0078] y[k]=C( k ) x ̇ (k)+v(k) ;

[0079] The discrete time Kalman filter is used to estimate the discrete optimal state vector by the following formula: x ^ [k|k] :

[0080] x ^ [k|k]= x ^ [k|k - 1]+ A ^ k K[k][y[k]- B ^ k P[k|k - 1] P k k = x ^ k|k - 1 + A ^ (k)P[k - 1|k - 1] B ^ T (k)+Q[k] ;

[0081] in, is the discrete state covariance matrix; Q[k] is the discrete process noise covariance matrix; K[k] is the discrete Kalman gain; then according to the estimated discrete optimal state vector x ^ [k|k] Estimate the discrete state transfer matrix and discrete control input matrix.

[0082] Specifically, the observation equation represents the observed value y [ k ] The state vector x [ k ] Observed Matrix Perform a linear transformation and add observation noise The obtained observation matrix Describes how to obtain observations from the state vector. In a clean environment management system, this can represent the environmental parameter measurements obtained from sensors. Discrete-time Kalman filtering is an estimation technique used to estimate the state of a system from observations. In this process, the estimated state vector x ˆ [ k ∣ k ] It can be used to monitor and control the environment. The following is the core formula of Kalman filtering:

[0083] x ˆ [ k ∣ k ]= x ˆ [ k ∣ k -1]+ A ˆ ( k ) K [ k ][ y [ k ]- B ˆ ( k ) P [ k ∣ k -1] ; This equation represents the discrete time step Next, the estimated state vector x ˆ [ k ∣ k ] is the state vector estimated by the previous step x ˆ [ k ∣ k -1] Adding the Kalman gain K [ k ] Multiply by the observed residual [ y [ k ]- B ˆ ( k ) P [ k ∣ k -1] The Kalman gain determines how to combine the observations with the estimates to get a more accurate state estimate. P [ k ∣ k ]= x ˆ [ k ∣ k -1]+ A ˆ ( k ) P [ k -1∣ k -1] B ˆ T ( k )+ Q [ k ] This equation represents the discrete time step Next, the state covariance matrix P [ k ∣ k ] will be updated to include information about the estimated error. P [ k -1∣ k -1] is the state covariance matrix of the previous step, and Q [ k ] is the process noise covariance matrix, which is used to represent the uncertainty and noise of the system. The application of these formulas can help the clean environment intelligent logistics management system to more accurately estimate environmental parameters, such as temperature, humidity, etc., from sensor observation data. By continuously updating the state estimate, the system can monitor the environment in real time and take control operations to maintain the desired environmental conditions. Kalman filtering technology can effectively handle observation noise and system uncertainty, and improve the performance and reliability of the environmental management system. In summary, this embodiment describes how to use Kalman filtering technology to estimate the state vector under discrete time, and how to use the observation equation and state covariance matrix for this process. This plays an important role in the clean environment intelligent logistics management system, enabling the system to respond to environmental changes in real time and improve the quality and stability of the working environment.

[0084] In the clean environment management system, one of the key tasks is to accurately estimate environmental parameters such as temperature, humidity, brightness, etc. Observation equation y [ k ]= C ( k ) x ˙ ( k )+ v ( k ) Allows the system to obtain real-time observations from sensors and measurement devices y [ k ] These observations are passed through the observation matrix The optimal state vector can be estimated by using a discrete time Kalman filter. x ˆ [ k ∣ k ] , that is, in discrete time steps This is one of the core functions of the clean environment management system, which allows the system to understand the current environmental conditions. x ˆ [ k ∣ k ] , the clean environment management system can monitor in real time whether the environmental parameters are consistent with the set values ​​or standards. If the environmental parameters deviate from the target, the system can take corresponding control operations to adjust the environmental conditions. These control operations can be represented by discrete-time state transfer matrices and discrete-time control input matrices, and the estimation of these matrices is usually also affected by discrete-time Kalman filtering. In practical applications, environmental management systems often face various uncertainties and noise sources, such as sensor errors, external interference, etc. Discrete-time Kalman filtering technology takes into account observation noise. and process noise Q [ k ] To handle these uncertainties. Observation noise reflects the error in sensor measurements, while process noise represents the uncertainty in the system model. In this way, the system can estimate the state of the environment more robustly. The discrete time Kalman filter not only provides a state estimate, but also a state covariance matrix P [ k ∣ k ] The matrix describes the uncertainty of the state estimate. This uncertainty information can be used for performance optimization and decision making. For example, the system can adjust the control strategy based on the uncertainty of the state estimate to ensure high-quality environmental management in an uncertain environment.

[0085] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A clean environment intelligent logistics management system based on digital twins, characterized by: The system includes: a clean environment intelligent logistics management model establishment unit, which is used to regard the clean environment intelligent logistics management system as an evolving system with a continuous time state and establish a corresponding abstract model, which specifically includes: within a set time period, collecting the operating parameters of the clean environment intelligent logistics management system under continuous time, and the operating parameters include: environmental quality parameters, control input parameters, control input actions, system response actions and noise; the environmental quality parameters are used as the state vector of the abstract model; the control input parameters are defined as the control parameter set issued by the clean environment intelligent logistics management system at a certain moment, and the control parameters are used as the control input vector of the abstract model; the system response action is defined as the action set of the clean environment intelligent logistics management system performing corresponding actions according to the data of the external environment perceived at a certain moment, and each action in the action set The elements are composed of numerical codes corresponding to the actions, and the system response action is used as the state transfer matrix of the abstract model; the control input action is defined as the set of actions received by the clean environment intelligent logistics management system at a certain moment, and the control input action is used as the control input matrix of the abstract model; the digital twin management and prediction unit is used to establish the corresponding observation equation according to the established abstract model, perform continuous time Kalman filtering, perform continuous time parameter estimation, and then perform continuous time model optimization. Finally, the continuous time is discretized to obtain the discrete time state equation, the observation equation is discretized, and the discrete time Kalman filter is used to estimate the discrete optimal state vector, and then estimate the discrete state transfer matrix and the discrete control input matrix. The clean environment intelligent logistics management system is controlled based on the estimated discrete state transfer matrix and discrete control input matrix; the abstract model is represented by the following formula: in, is the abstract model variable, x(t) is the state vector; A(t) is the state transfer matrix; B(t) is the control input matrix; u(t) is the control input vector; represents the Laplace operation of the state vector x(t); represents the control input integrated over time, representing the cumulative effect of the control input over time; s is the integral variable; u(s) is the control input vector at time s; B(s) is the control input matrix at time s; Ω(t) is the noise, representing the nonlinear and time-varying system disturbance.

2. The clean environment intelligent logistics management system based on digital twin according to claim 1, characterized in that: The environmental quality parameters include at least: temperature, humidity, pressure difference, air quality, light brightness and noise level; the control input parameters include at least: temperature setting value, humidity setting value, wind speed, wind direction and light brightness setting value.

3. The clean environment intelligent logistics management system based on digital twin according to claim 2, characterized in that: The control input actions include at least: temperature increase, temperature decrease, humidity increase, humidity decrease, wind speed adjustment, wind direction adjustment, light brightness increase, light brightness decrease, air purification device turned on and air purification device turned off; the system response actions include at least: temperature increase, temperature decrease, humidity increase, humidity decrease, wind speed adjustment, wind direction adjustment, light brightness increase, light brightness decrease, air purification device turned on and air purification device turned off.

4. The clean environment intelligent logistics management system based on digital twin according to claim 3 is characterized in that: The observation equation is expressed as follows: Among them, y(t) is the observation vector; C(t) is the observation matrix; v(t) is the observation noise.

5. The clean environment intelligent logistics management system based on digital twin according to claim 4, characterized in that: Use the following formula to perform continuous time Kalman filtering to obtain the optimal state estimate vector: in, is the optimal state estimate vector; is the optimal state estimation vector at the previous moment; K(t) is the Kalman gain.

6. The clean environment intelligent logistics management system based on digital twin according to claim 5, characterized in that: The method for continuous time parameter estimation includes: defining a loss function of a state transfer matrix A(t) and a control input matrix B(t); obtaining an optimization problem of parameter estimation by minimizing the loss function; estimating the state transfer matrix and the control input matrix by solving the optimization problem of parameter estimation; the loss function is expressed as: in, is the loss function value; by minimizing the loss function, the optimization problem of parameter estimation is expressed using the following formula: is the estimated state transfer matrix; To estimate the control input matrix, use the following formula to solve the optimization problem of parameter estimation: based on and The continuous time model is optimized, and the optimized abstract model is:

7. The clean environment intelligent logistics management system based on digital twin according to claim 6, characterized in that: The optimized abstract model is discretized in continuous time, and the discrete time state equation is expressed using the following formula: Where k is a discrete time variable; is a discrete integral variable.

8. The clean environment intelligent logistics management system based on digital twin according to claim 7, characterized in that: The observation equation is discretized by the following formula: The discrete time Kalman filter is used to estimate the discrete optimal state vector by the following formula: Where P[k|k] is the discrete state covariance matrix; Q[k] is the discrete process noise covariance matrix; K[k] is the discrete Kalman gain; and then according to the estimated discrete optimal state vector Estimate the discrete state transfer matrix and discrete control input matrix.

Images