Dynamic optimization system and method for ratio of compound essential oil

Abstract

The invention discloses a dynamic optimization system and method for a compound essential oil ratio, and aims to solve the problems that the traditional technology depends on artificial experience, the optimization efficiency is low and the quality tracing is insufficient. According to the method, raw material molecular fingerprint data are collected in real time through a high-precision spectrum sensor array and a micro-fluidic chip, a production environment field domain is constructed in combination with a computational fluid mechanics model, a formula optimization problem is encoded into an Ising model by using a quantum annealing algorithm, and global search is realized through a quantum tunneling effect. The system integrates digital twin bodies, covers molecular dynamics simulation, phase equilibrium prediction and olfactory receptor activation probability simulation, and accurately maps a physical system state; a multi-modal data fusion network and a reinforcement learning framework are adopted to dynamically adjust formula parameters, and an NSGA-III algorithm is combined to optimize efficacy collaboration degree, cost and stability indexes. And the execution control module realizes nanoscale flow regulation through the piezoelectric micro-injection valve array, and realizes non-tampering evidence storage of formula parameters and quality data based on a block chain technology. Dynamic closed-loop optimization of the compound essential oil ratio is achieved, the production efficiency, the product consistency and the supply chain transparency are remarkably improved, and the method is suitable for the field of intelligent manufacturing of high-end essential oil.

Description

Technical Field

[0001] The present invention relates to the technical field of compound essential oil ratios, in particular to a dynamic optimization system and method for compound essential oil ratios. Background Art

[0002] Existing blending technologies for essential oil blends often rely on manual experience or static optimization models, which present significant limitations. First, traditional data acquisition methods (such as offline spectral analysis or single-sensor detection) struggle to capture multidimensional molecular fingerprint information (e.g., UV-visible spectra and dynamic viscosity changes) of raw materials in real time. This results in delayed blending adjustments and an inability to adapt to transient fluctuations in the production environment. Second, optimization algorithms often employ classical heuristic methods (such as genetic algorithms or gradient descent), which are prone to local optima in complex multi-objective problems and struggle to strike a global balance between efficacy synergy, cost, and stability. Furthermore, the production process lacks a dynamic closed-loop control mechanism, and fixed blending parameters cannot effectively respond to heat and mass transfer coupling effects or changes in volatilization dynamics, impacting product consistency. Existing digital simulation models are often disconnected from the physical system and fail to integrate molecular dynamics, phase equilibrium prediction, and olfactory receptor activation simulation, resulting in insufficient prediction accuracy. Furthermore, quality data traceability relies on centralized databases, which pose a risk of tampering and lack supply chain transparency, making it difficult to meet the product credibility requirements of the high-end market. These issues hinder the efficiency, intelligence, and traceability of essential oil blend production. Summary of the Invention

[0003] In order to solve the above technical problems, the technical solution adopted by the dynamic optimization method of the compound essential oil ratio of the present invention comprises the following steps: Step 1: Real-time collection of raw material molecular fingerprint data, including UV-visible spectra, near-infrared spectra, and viscosity parameters, using a high-precision spectral sensor array and microfluidic chip; Step 2: Establish a production environment field model based on computational fluid dynamics, integrating the mass transfer-heat transfer coupling equation and the volatilization kinetic parameter prediction module; Step 3: Use the quantum annealing algorithm to encode the recipe combination optimization problem into an Ising model, and realize the global search of the recipe space through the quantum tunneling effect; Step 4: Build a digital twin that includes molecular dynamics simulation and phase equilibrium prediction to map the multi-scale state of the physical production system in real time; Step 5: Integrate sensor data and virtual simulation data through a multimodal data fusion network to generate a recipe feature latent space vector; Step 6: Dynamically adjust the recipe parameters based on the reinforcement learning framework and define the state space of the Markov decision process as a combination of real-time sensor data and environmental parameters; Step 7: Use the multi-objective evolutionary algorithm NSGA-III to optimize the efficacy synergy, cost and stability indicators in the three-dimensional target space; Step 8: Perform nanoscale flow control through a piezoelectric micro-injection valve array and dynamically adjust PID parameters in combination with the Lyapunov stability criterion; Step 9: Build a blockchain evidence chain based on Hyperledger Fabric to achieve tamper-proof traceability of formula parameters and quality data.

[0004] As a further solution of the present invention, the spectral sensor array in step 1 includes a UV-Vis / NIR / MIR three-band synchronous detection module, a microfluidic chip integrated with a surface tension sensing unit, and a sampling frequency of not less than 100 Hz.

[0005] As a further solution of the present invention, the quantum optimization process in step 3 includes: S31: Encoding the compatibility relationship of essential oil components into the spin glass model Hamiltonian; S32: Solve the formula combination corresponding to the ground state energy through the D-Wave quantum annealer; S33: Using quantum tunneling effect to avoid the local optimal trap of traditional optimization algorithms.

[0006] As a further solution of the present invention, the digital twin in step 4 includes: S41: Intermolecular interaction energy matrix based on AMBER force field; S42: Phase equilibrium predictor combined with UNIFAC-PR model; S43: Olfactory receptor activation probability calculation module, hOR17-4 protein molecular docking simulation.

[0007] As a further solution of the present invention, the multimodal data fusion in step 5 adopts a hybrid architecture of spatiotemporal feature pyramid network (3D-CNN) and Transformer attention mechanism, and generates a 128-dimensional latent space feature vector through variational autoencoder (VAE).

[0008] As a further solution of the present invention, the reinforcement learning framework in step 6 adopts the double-delayed deep deterministic policy gradient (TD3) algorithm, and the reward function is defined as: , in As a real-time efficacy indicator, is the physical property parameter, weight coefficient , Dynamic adaptive adjustment.

[0009] As a further solution of the present invention, the multi-objective optimization in step 7 includes a reference vector-guided Pareto frontier search strategy, adopts an elite retention mechanism based on crowding distance, and the optimization objective function includes: S71: Efficacy synergy (CI index): , where i and j represent different components in the compound essential oil. It indicates the synergistic effect when two ingredients work together. and represents the half-maximal effect concentration of components i and j, respectively, i.e., the concentration required to achieve 50% of the maximum efficacy when used alone; S72: Cost function: weighted sum of raw material purchase price and energy consumption; S73: Stability index: Standard deviation of ingredient retention under accelerated test.

[0010] As a further solution of the present invention, the execution control module in step 8 includes: S81: Nanoscale flow feedback control of piezoelectric microjet valve array; S82: PID parameter tuning rules based on online calculation of Lyapunov exponents; S83: Minimum entropy production rate control strategy for non-equilibrium processes.

[0011] As a further solution of the present invention, the blockchain evidence chain in step 9 includes a smart contract verification module, which specifically implements: S91: SHA-3 hash values ​​of recipe parameters and production batch data are uploaded to the blockchain; S92: Quality verification protocol based on zero-knowledge proof; S93: Implement multi-level supply chain traceability through Merkle tree structure.

[0012] As a further embodiment of the present invention, a dynamic optimization system for compound essential oil ratios is characterized by comprising: Data acquisition module: spectral sensor array, electronic nose system, microfluidic chip; Digital twin building blocks: molecular dynamics simulation engine, phase equilibrium predictor; Intelligent optimization module: quantum-classical hybrid optimizer, multi-objective evolutionary algorithm core; Execution control module: piezoelectric micro-injection actuator, self-correcting PID controller and blockchain evidence storage platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The present invention provides a flow chart of the steps of a dynamic optimization method for compound essential oil ratios. DETAILED DESCRIPTION

[0014] The present invention will be further described in detail below with reference to the embodiments.

[0015] See also Figure 1 FIG. 5 is a flow chart of a method for dynamically optimizing a compound essential oil ratio according to the present invention, wherein the specific steps include: Step 1: Real-time collection of raw material molecular fingerprint data, including UV-visible spectra, near-infrared spectra, and viscosity parameters, using a high-precision spectral sensor array and microfluidic chip; Step 2: Establish a production environment field model based on computational fluid dynamics, integrating the mass transfer-heat transfer coupling equation and the volatilization kinetic parameter prediction module; Step 3: Use the quantum annealing algorithm to encode the recipe combination optimization problem into an Ising model, and realize the global search of the recipe space through the quantum tunneling effect; Step 4: Build a digital twin that includes molecular dynamics simulation and phase equilibrium prediction to map the multi-scale state of the physical production system in real time; Step 5: Integrate sensor data and virtual simulation data through a multimodal data fusion network to generate a recipe feature latent space vector; Step 6: Dynamically adjust the recipe parameters based on the reinforcement learning framework and define the state space of the Markov decision process as a combination of real-time sensor data and environmental parameters; Step 7: Use the multi-objective evolutionary algorithm NSGA-III to optimize the efficacy synergy, cost and stability indicators in the three-dimensional target space; Step 8: Perform nanoscale flow control through a piezoelectric micro-injection valve array and dynamically adjust PID parameters in combination with the Lyapunov stability criterion; Step 9: Build a blockchain evidence chain based on Hyperledger Fabric to achieve tamper-proof traceability of formula parameters and quality data.

[0016] Furthermore, the spectral sensor array in step 1 includes a UV-Vis / NIR / MIR three-band synchronous detection module, a microfluidic chip integrated with a surface tension sensing unit, and a sampling frequency of not less than 100 Hz.

[0017] Furthermore, the quantum optimization process in step 3 includes: S31: Encoding the compatibility relationship of essential oil components into the spin glass model Hamiltonian; S32: Solve the formula combination corresponding to the ground state energy through the D-Wave quantum annealer; S33: Using quantum tunneling effect to avoid the local optimal trap of traditional optimization algorithms.

[0018] Furthermore, the digital twin in step 4 includes: S41: Intermolecular interaction energy matrix based on AMBER force field; S42: Phase equilibrium predictor combined with UNIFAC-PR model; S43: Olfactory receptor activation probability calculation module, hOR17-4 protein molecular docking simulation.

[0019] Furthermore, in step 5, the multimodal data fusion adopts a hybrid architecture of spatiotemporal feature pyramid network (3D-CNN) and Transformer attention mechanism, and generates a 128-dimensional latent space feature vector through variational autoencoder (VAE).

[0020] Furthermore, the reinforcement learning framework in step 6 adopts the double-delayed deep deterministic policy gradient (TD3) algorithm, and the reward function is defined as:

[0021] in As a real-time efficacy indicator, is the physical property parameter, weight coefficient , Dynamic adaptive adjustment.

[0022] Furthermore, the multi-objective optimization in step 7 includes a reference vector-guided Pareto frontier search strategy, adopting an elite retention mechanism based on crowding distance, and the optimization objective function includes: S71: Efficacy synergy (CI index): , where i and j represent different components in the compound essential oil. It indicates the synergistic effect when two ingredients work together. and represents the half-maximal effect concentration of components i and j, respectively, i.e., the concentration required to achieve 50% of the maximum efficacy when used alone; S72: Cost function: weighted sum of raw material purchase price and energy consumption; S73: Stability index: Standard deviation of ingredient retention under accelerated test.

[0023] Furthermore, the execution control module in step 8 includes: S81: Nanoscale flow feedback control of piezoelectric microjet valve array; S82: PID parameter tuning rules based on online calculation of Lyapunov exponents; S83: Minimum entropy production rate control strategy for non-equilibrium processes.

[0024] Furthermore, in step 9, the blockchain evidence chain includes a smart contract verification module, which specifically implements: S91: SHA-3 hash values ​​of recipe parameters and production batch data are uploaded to the blockchain; S92: Quality verification protocol based on zero-knowledge proof; S93: Implement multi-level supply chain traceability through Merkle tree structure.

[0025] Furthermore, a dynamic optimization system for compound essential oil ratios is characterized by comprising: Data acquisition module: spectral sensor array, electronic nose system, microfluidic chip; Digital twin building blocks: molecular dynamics simulation engine, phase equilibrium predictor; Intelligent optimization module: quantum-classical hybrid optimizer, multi-objective evolutionary algorithm core; Execution control module: piezoelectric micro-injection actuator, self-correcting PID controller and blockchain evidence storage platform.

[0026] The foregoing is merely an embodiment of the present invention and is not intended to limit the present invention. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims of this application.

Claims

1. A dynamic optimization method for compound essential oil ratios, characterized in that, The following steps are involved: Step 1: Real-time collection of raw material molecular fingerprint data, including UV-visible spectra, near-infrared spectra, and viscosity parameters, using a high-precision spectral sensor array and microfluidic chip; Step 2: Establish a production environment field model based on computational fluid dynamics, integrating the mass transfer-heat transfer coupling equation and the volatilization kinetic parameter prediction module; Step 3: Use the quantum annealing algorithm to encode the recipe combination optimization problem into an Ising model, and realize the global search of the recipe space through the quantum tunneling effect; Step 4: Build a digital twin that includes molecular dynamics simulation and phase equilibrium prediction to map the multi-scale state of the physical production system in real time; Step 5: Integrate sensor data and virtual simulation data through a multimodal data fusion network to generate a recipe feature latent space vector; Step 6: Dynamically adjust the recipe parameters based on the reinforcement learning framework and define the state space of the Markov decision process as a combination of real-time sensor data and environmental parameters; Step 7: Use the multi-objective evolutionary algorithm NSGA-III to optimize the efficacy synergy, cost and stability indicators in the three-dimensional target space; Step 8: Perform nanoscale flow control through a piezoelectric micro-injection valve array and dynamically adjust PID parameters in combination with the Lyapunov stability criterion; Step 9: Build a blockchain evidence chain based on Hyperledger Fabric to achieve tamper-proof traceability of formula parameters and quality data.

2. the dynamic optimization method of a kind of compound essential oil proportioning according to claim 1, is characterized in that: The spectral sensor array in step 1 includes a UV-Vis / NIR / MIR three-band synchronous detection module, a microfluidic chip integrated with a surface tension sensing unit, and a sampling frequency of not less than 100 Hz.

3. the dynamic optimization method of a kind of compound essential oil proportioning according to claim 1, is characterized in that: The quantum optimization process in step 3 includes: S31: Encoding the compatibility relationship of essential oil components into the spin glass model Hamiltonian; S32: Solve the formula combination corresponding to the ground state energy through the D-Wave quantum annealer; S33: Using quantum tunneling effect to avoid the local optimal trap of traditional optimization algorithms.

4. the dynamic optimization method of a kind of compound essential oil proportioning according to claim 1, is characterized in that: The digital twin in step 4 includes: S41: Intermolecular interaction energy matrix based on AMBER force field; S42: Phase equilibrium predictor combined with UNIFAC-PR model; S43: Olfactory receptor activation probability calculation module, hOR17-4 protein molecular docking simulation.

5. the dynamic optimization method of a kind of compound essential oil proportioning according to claim 1, is characterized in that: The multimodal data fusion in step 5 adopts a hybrid architecture of the spatiotemporal feature pyramid network and the Transformer attention mechanism, and generates a 128-dimensional latent space feature vector through a variational autoencoder.

6. a dynamic optimization method for a compound essential oil proportioning according to claim 1, is characterized in that: The reinforcement learning framework in step 6 adopts a double-delayed deep deterministic policy gradient algorithm, and the reward function is defined as: ,in As a real-time efficacy indicator, is the physical property parameter, weight coefficient , Dynamic adaptive adjustment.

7. a dynamic optimization method for a compound essential oil proportioning according to claim 1, is characterized in that: The multi-objective optimization in step 7 includes a reference vector-guided Pareto frontier search strategy, adopts an elite retention mechanism based on crowding distance, and the optimization objective function includes: S71: Efficacy synergy (CI index): , where i and j represent different components in the compound essential oil. It indicates the synergistic effect when two ingredients work together. and represents the half-maximal effect concentration of components i and j, respectively, i.e., the concentration required to achieve 50% of the maximum efficacy when used alone; S72: Cost function: weighted sum of raw material purchase price and energy consumption; S73: Stability index: Standard deviation of ingredient retention under accelerated test.

8. a dynamic optimization method for a compound essential oil proportioning according to claim 1, is characterized in that: The execution control module in step 8 includes: S81: Nanoscale flow feedback control of piezoelectric microjet valve array; S82: PID parameter tuning rules based on online calculation of Lyapunov exponents; S83: Minimum entropy production rate control strategy for non-equilibrium processes.

9. a dynamic optimization method for a compound essential oil proportioning according to claim 1, is characterized in that: In step 9, the blockchain evidence chain includes a smart contract verification module, which is specifically implemented as follows: S91: SHA-3 hash values ​​of recipe parameters and production batch data are uploaded to the blockchain; S92: Quality verification protocol based on zero-knowledge proof; S93: Implement multi-level supply chain traceability through Merkle tree structure.

10. A dynamic optimization system for implementing the method according to any one of claims 1 to 9, characterized in that include: Data acquisition module: spectral sensor array, electronic nose system, microfluidic chip; Digital twin building blocks: molecular dynamics simulation engine, phase equilibrium predictor; Intelligent optimization module: quantum-classical hybrid optimizer, multi-objective evolutionary algorithm core; Execution control module: piezoelectric micro-injection actuator, self-correcting PID controller and blockchain evidence storage platform.

Images