Adaptive Differentiating Engine for High-Rate Signal Derivatives
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for real-time derivative estimation of analog and digital signals, particularly at high sampling rates, are prone to noise and systematic errors due to introduced time delays, making them unsuitable for high precision control systems.
Innovation Solution
A differentiating engine that employs sliding mode control and adaptive gain methods to calculate the derivative of input signals, using pre-determined gains and error estimation to output accurate, noise-free signal derivatives in real-time, suitable for high sampling rate applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing differentiator systems are used for real-time derivative estimation at high sampling rates, then the system can operate at high sampling rates, but the output is prone to noise and systematic error
Solution Approach 1:
The patent implements a differentiator with adaptive gain scheduling that dynamically adjusts the differentiation parameter based on the input signal frequency. The system continuously monitors the input signal characteristics and modifies the differentiation gain in real-time, transitioning from fixed-parameter differentiators to adaptive ones. This dynamic adjustment allows the system to maintain high sampling rates while compensating for frequency-dependent errors, thereby resolving the contradiction between high productivity and measurement precision.
Solution Approach 2:
The patent changes the differentiation parameter (gain) as a function of the input signal frequency. By establishing a relationship between the differentiation parameter and the spectral content of the input signal, the system optimizes the differentiation accuracy for different operating conditions. This parameter adaptation enables the differentiator to maintain precision across varying sampling rates and signal characteristics, directly addressing the accuracy degradation at high sampling rates.
2Measurement precision
If low noise differentiator systems are used to reduce noise and systematic error, then the measurement precision is improved, but unacceptable time delays are introduced into the control system
Solution Approach 1:
The patent employs a dynamic differentiator where the differentiation parameter is continuously adapted based on the input signal frequency rather than using fixed conservative parameters. This dynamic approach allows the system to achieve low noise performance without the excessive time delays associated with traditional low-noise differentiators that use fixed high-order filtering. The adaptive mechanism optimizes the trade-off between noise reduction and time delay in real-time.
Solution Approach 2:
The patent performs preliminary frequency analysis of the input signal to determine the appropriate differentiation parameter before computing the derivative. By pre-assessing the signal characteristics and selecting optimal parameters in advance, the system avoids the need for excessive post-processing filtering that would introduce time delays. This preliminary action enables accurate derivative estimation with minimal latency.
3Productivity
If high sampling rates are used to improve real-time control responsiveness, then the productivity is improved, but noise and systematic error increase
Solution Approach 1:
The patent modifies the differentiation parameter based on the input signal frequency to optimize performance at high sampling rates. By establishing an explicit relationship between the sampling rate, signal frequency, and differentiation gain, the system compensates for the increased noise and systematic errors inherent in high-rate sampling. This parameter adaptation allows the system to maintain high productivity while suppressing harmful factors through mathematically optimized differentiation.
Data Source
AI summary
The controller includes a differentiating engine configured to receive an input signal value (ISV), wherein the ISV corresponds to state information for one selected from a group consisting of a controlled process and a user interface. The differentiating engine is further configured to determine an error between the ISV and an estimated input signal (EIS), estimate a frequency of the IS, select a plurality of pre-determined gains using the frequency, wherein at least one plurality of pre-determined gains is a suction control gain, determine a first estimated derivative of the input signal (EDIS) using the plurality of pre-determined gains and the error, and to output the first EDIS.